Nomad changes

packages/java/dev/kreuzberg/AccelerationConfig.java

@@ -0,0 +1,67 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Hardware acceleration configuration for ONNX Runtime models.
+ *
+ * Controls which execution provider (CPU, CoreML, CUDA, TensorRT) is used
+ * for inference in layout detection and embedding generation.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = AccelerationConfig.Builder.class)
+public record AccelerationConfig(
+    /**
+     * Execution provider to use for ONNX inference.
+     */
+    @Nullable @JsonProperty("provider") ExecutionProviderType provider,
+    /**
+     * GPU device ID (for CUDA/TensorRT). Ignored for CPU/CoreML/Auto.
+     */
+    @Nullable @JsonProperty("device_id") Integer deviceId
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @Nullable private ExecutionProviderType provider = ExecutionProviderType.Auto;
+        @JsonProperty("device_id")
+private Integer deviceId = null;
+
+        /** Sets the provider field. */
+        @JsonProperty("provider")
+        public Builder withProvider(final @Nullable ExecutionProviderType value) {
+            this.provider = value;
+            return this;
+        }
+
+        /** Sets the deviceId field. */
+        @JsonProperty("device_id")
+        public Builder withDeviceId(final @Nullable Integer value) {
+            this.deviceId = value;
+            return this;
+        }
+
+        /** Builds the AccelerationConfig instance. */
+        public AccelerationConfig build() {
+            return new AccelerationConfig(
+                provider,
+                deviceId
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/AnnotationKind.java

@@ -0,0 +1,77 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonSubTypes;
+import com.fasterxml.jackson.annotation.JsonTypeInfo;
+import java.util.Optional;
+
+/**
+ * Types of inline text annotations.
+ */
+@JsonTypeInfo(use = JsonTypeInfo.Id.NAME, property = "annotation_type", visible = false)
+@JsonSubTypes({
+    @JsonSubTypes.Type(value = AnnotationKind.Bold.class, name = "bold"),
+    @JsonSubTypes.Type(value = AnnotationKind.Italic.class, name = "italic"),
+    @JsonSubTypes.Type(value = AnnotationKind.Underline.class, name = "underline"),
+    @JsonSubTypes.Type(value = AnnotationKind.Strikethrough.class, name = "strikethrough"),
+    @JsonSubTypes.Type(value = AnnotationKind.Code.class, name = "code"),
+    @JsonSubTypes.Type(value = AnnotationKind.Subscript.class, name = "subscript"),
+    @JsonSubTypes.Type(value = AnnotationKind.Superscript.class, name = "superscript"),
+    @JsonSubTypes.Type(value = AnnotationKind.Link.class, name = "link"),
+    @JsonSubTypes.Type(value = AnnotationKind.Highlight.class, name = "highlight"),
+    @JsonSubTypes.Type(value = AnnotationKind.Color.class, name = "color"),
+    @JsonSubTypes.Type(value = AnnotationKind.FontSize.class, name = "font_size"),
+    @JsonSubTypes.Type(value = AnnotationKind.Custom.class, name = "custom")
+})
+@com.fasterxml.jackson.annotation.JsonIgnoreProperties(ignoreUnknown = true)
+public sealed interface AnnotationKind {
+
+    record Bold() implements AnnotationKind {
+    }
+
+    record Italic() implements AnnotationKind {
+    }
+
+    record Underline() implements AnnotationKind {
+    }
+
+    record Strikethrough() implements AnnotationKind {
+    }
+
+    record Code() implements AnnotationKind {
+    }
+
+    record Subscript() implements AnnotationKind {
+    }
+
+    record Superscript() implements AnnotationKind {
+    }
+
+    record Link(
+        @JsonProperty("url") String url,
+        @JsonProperty("title") Optional<String> title
+    ) implements AnnotationKind {
+    }
+
+    /** Highlighted text (PDF highlights, HTML {@code &lt;mark&gt;}). */
+    record Highlight() implements AnnotationKind {
+    }
+
+    /** Text color (CSS-compatible value, e.g. "#ff0000", "red"). */
+    record Color(@JsonProperty("value") String value) implements AnnotationKind { }
+
+    /** Font size with units (e.g. "12pt", "1.2em", "16px"). */
+    record FontSize(@JsonProperty("value") String value) implements AnnotationKind { }
+
+    /** Extensible annotation for format-specific styling. */
+    record Custom(
+        @JsonProperty("name") String name,
+        @JsonProperty("value") Optional<String> value
+    ) implements AnnotationKind {
+    }
+}

packages/java/dev/kreuzberg/ArchiveEntry.java

@@ -0,0 +1,79 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * A single file extracted from an archive.
+ *
+ * When archives (ZIP, TAR, 7Z, GZIP) are extracted with recursive extraction
+ * enabled, each processable file produces its own full {@code ExtractionResult}.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ArchiveEntry.Builder.class)
+public record ArchiveEntry(
+    /**
+     * Archive-relative file path (e.g. "folder/document.pdf").
+     */
+    @JsonProperty("path") String path,
+    /**
+     * Detected MIME type of the file.
+     */
+    @JsonProperty("mime_type") String mimeType,
+    /**
+     * Full extraction result for this file.
+     */
+    @JsonProperty("result") ExtractionResult result
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String path = "";
+        @JsonProperty("mime_type")
+private String mimeType = "";
+private ExtractionResult result = null;
+
+        /** Sets the path field. */
+        @JsonProperty("path")
+        public Builder withPath(final String value) {
+            this.path = value;
+            return this;
+        }
+
+        /** Sets the mimeType field. */
+        @JsonProperty("mime_type")
+        public Builder withMimeType(final String value) {
+            this.mimeType = value;
+            return this;
+        }
+
+        /** Sets the result field. */
+        @JsonProperty("result")
+        public Builder withResult(final ExtractionResult value) {
+            this.result = value;
+            return this;
+        }
+
+        /** Builds the ArchiveEntry instance. */
+        public ArchiveEntry build() {
+            return new ArchiveEntry(
+                path,
+                mimeType,
+                result
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/ArchiveMetadata.java

@@ -0,0 +1,109 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Archive (ZIP/TAR/7Z) metadata.
+ *
+ * Extracted from compressed archive files containing file lists and size information.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ArchiveMetadata.Builder.class)
+public record ArchiveMetadata(
+    /**
+     * Archive format ("ZIP", "TAR", "7Z", etc.)
+     */
+    @JsonProperty("format") String format,
+    /**
+     * Total number of files in the archive
+     */
+    @JsonProperty("file_count") int fileCount,
+    /**
+     * List of file paths within the archive
+     */
+    @JsonProperty("file_list") List<String> fileList,
+    /**
+     * Total uncompressed size in bytes
+     */
+    @JsonProperty("total_size") long totalSize,
+    /**
+     * Compressed size in bytes (if available)
+     */
+    @Nullable @JsonProperty("compressed_size") Long compressedSize
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String format = "";
+        @JsonProperty("file_count")
+private int fileCount = 0;
+        @JsonProperty("file_list")
+private List<String> fileList = List.of();
+        @JsonProperty("total_size")
+private long totalSize = 0;
+        @JsonProperty("compressed_size")
+private Long compressedSize = null;
+
+        /** Sets the format field. */
+        @JsonProperty("format")
+        public Builder withFormat(final String value) {
+            this.format = value;
+            return this;
+        }
+
+        /** Sets the fileCount field. */
+        @JsonProperty("file_count")
+        public Builder withFileCount(final int value) {
+            this.fileCount = value;
+            return this;
+        }
+
+        /** Sets the fileList field. */
+        @JsonProperty("file_list")
+        public Builder withFileList(final List<String> value) {
+            this.fileList = value;
+            return this;
+        }
+
+        /** Sets the totalSize field. */
+        @JsonProperty("total_size")
+        public Builder withTotalSize(final long value) {
+            this.totalSize = value;
+            return this;
+        }
+
+        /** Sets the compressedSize field. */
+        @JsonProperty("compressed_size")
+        public Builder withCompressedSize(final @Nullable long value) {
+            this.compressedSize = value;
+            return this;
+        }
+
+        /** Builds the ArchiveMetadata instance. */
+        public ArchiveMetadata build() {
+            return new ArchiveMetadata(
+                format,
+                fileCount,
+                fileList,
+                totalSize,
+                compressedSize
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/BBox.java

@@ -0,0 +1,76 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * Bounding box in original image coordinates (x1, y1) top-left, (x2, y2) bottom-right.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = BBox.Builder.class)
+public record BBox(
+    @JsonProperty("x1") float x1,
+    @JsonProperty("y1") float y1,
+    @JsonProperty("x2") float x2,
+    @JsonProperty("y2") float y2
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private float x1 = 0.0f;
+private float y1 = 0.0f;
+private float x2 = 0.0f;
+private float y2 = 0.0f;
+
+        /** Sets the x1 field. */
+        @JsonProperty("x1")
+        public Builder withX1(final float value) {
+            this.x1 = value;
+            return this;
+        }
+
+        /** Sets the y1 field. */
+        @JsonProperty("y1")
+        public Builder withY1(final float value) {
+            this.y1 = value;
+            return this;
+        }
+
+        /** Sets the x2 field. */
+        @JsonProperty("x2")
+        public Builder withX2(final float value) {
+            this.x2 = value;
+            return this;
+        }
+
+        /** Sets the y2 field. */
+        @JsonProperty("y2")
+        public Builder withY2(final float value) {
+            this.y2 = value;
+            return this;
+        }
+
+        /** Builds the BBox instance. */
+        public BBox build() {
+            return new BBox(
+                x1,
+                y1,
+                x2,
+                y2
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/BatchBytesItem.java

@@ -0,0 +1,81 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonSerialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Batch item for byte array extraction.
+ *
+ * Used with {@code batch_extract_bytes} and {@code batch_extract_bytes_sync}
+ * to represent a single item in a batch extraction job.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = BatchBytesItem.Builder.class)
+public record BatchBytesItem(
+    /**
+     * The content bytes to extract from
+     */
+    @JsonSerialize(using = ByteArrayToIntArraySerializer.class) @JsonProperty("content") byte[] content,
+    /**
+     * MIME type of the content (e.g., "application/pdf", "text/html")
+     */
+    @JsonProperty("mime_type") String mimeType,
+    /**
+     * Per-item configuration overrides (null uses batch-level defaults)
+     */
+    @Nullable @JsonProperty("config") FileExtractionConfig config
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private byte[] content = new byte[0];
+        @JsonProperty("mime_type")
+private String mimeType = "";
+private FileExtractionConfig config = null;
+
+        /** Sets the content field. */
+        @JsonProperty("content")
+        public Builder withContent(final byte[] value) {
+            this.content = value;
+            return this;
+        }
+
+        /** Sets the mimeType field. */
+        @JsonProperty("mime_type")
+        public Builder withMimeType(final String value) {
+            this.mimeType = value;
+            return this;
+        }
+
+        /** Sets the config field. */
+        @JsonProperty("config")
+        public Builder withConfig(final @Nullable FileExtractionConfig value) {
+            this.config = value;
+            return this;
+        }
+
+        /** Builds the BatchBytesItem instance. */
+        public BatchBytesItem build() {
+            return new BatchBytesItem(
+                content,
+                mimeType,
+                config
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/BatchFileItem.java

@@ -0,0 +1,66 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Batch item for file extraction.
+ *
+ * Used with {@code batch_extract_files} and {@code batch_extract_files_sync}
+ * to represent a single file in a batch extraction job.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = BatchFileItem.Builder.class)
+public record BatchFileItem(
+    /**
+     * Path to the file to extract from
+     */
+    @JsonProperty("path") java.nio.file.Path path,
+    /**
+     * Per-file configuration overrides (null uses batch-level defaults)
+     */
+    @Nullable @JsonProperty("config") FileExtractionConfig config
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private java.nio.file.Path path = null;
+private FileExtractionConfig config = null;
+
+        /** Sets the path field. */
+        @JsonProperty("path")
+        public Builder withPath(final java.nio.file.Path value) {
+            this.path = value;
+            return this;
+        }
+
+        /** Sets the config field. */
+        @JsonProperty("config")
+        public Builder withConfig(final @Nullable FileExtractionConfig value) {
+            this.config = value;
+            return this;
+        }
+
+        /** Builds the BatchFileItem instance. */
+        public BatchFileItem build() {
+            return new BatchFileItem(
+                path,
+                config
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/BibtexMetadata.java

@@ -0,0 +1,96 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import java.util.Map;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * BibTeX bibliography metadata.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = BibtexMetadata.Builder.class)
+public record BibtexMetadata(
+    /**
+     * Number of entries in the bibliography.
+     */
+    @JsonProperty("entry_count") long entryCount,
+    @Nullable @JsonProperty("citation_keys") List<String> citationKeys,
+    @Nullable @JsonProperty("authors") List<String> authors,
+    @Nullable @JsonProperty("year_range") YearRange yearRange,
+    @Nullable @JsonProperty("entry_types") Map<String, Long> entryTypes
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("entry_count")
+private long entryCount = 0;
+        @JsonProperty("citation_keys")
+private List<String> citationKeys = null;
+private List<String> authors = null;
+        @JsonProperty("year_range")
+private YearRange yearRange = null;
+        @JsonProperty("entry_types")
+private Map<String, Long> entryTypes = null;
+
+        /** Sets the entryCount field. */
+        @JsonProperty("entry_count")
+        public Builder withEntryCount(final long value) {
+            this.entryCount = value;
+            return this;
+        }
+
+        /** Sets the citationKeys field. */
+        @JsonProperty("citation_keys")
+        public Builder withCitationKeys(final @Nullable List<String> value) {
+            this.citationKeys = value;
+            return this;
+        }
+
+        /** Sets the authors field. */
+        @JsonProperty("authors")
+        public Builder withAuthors(final @Nullable List<String> value) {
+            this.authors = value;
+            return this;
+        }
+
+        /** Sets the yearRange field. */
+        @JsonProperty("year_range")
+        public Builder withYearRange(final @Nullable YearRange value) {
+            this.yearRange = value;
+            return this;
+        }
+
+        /** Sets the entryTypes field. */
+        @JsonProperty("entry_types")
+        public Builder withEntryTypes(final @Nullable Map<String, Long> value) {
+            this.entryTypes = value;
+            return this;
+        }
+
+        /** Builds the BibtexMetadata instance. */
+        public BibtexMetadata build() {
+            return new BibtexMetadata(
+                entryCount,
+                citationKeys,
+                authors,
+                yearRange,
+                entryTypes
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/BlockType.java

@@ -0,0 +1,61 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonCreator;
+import com.fasterxml.jackson.annotation.JsonValue;
+
+/**
+ * Types of block-level elements in Djot.
+ */
+public enum BlockType {
+    Paragraph("paragraph"),
+    Heading("heading"),
+    Blockquote("blockquote"),
+    CodeBlock("code_block"),
+    ListItem("list_item"),
+    OrderedList("ordered_list"),
+    BulletList("bullet_list"),
+    TaskList("task_list"),
+    DefinitionList("definition_list"),
+    DefinitionTerm("definition_term"),
+    DefinitionDescription("definition_description"),
+    Div("div"),
+    Section("section"),
+    ThematicBreak("thematic_break"),
+    RawBlock("raw_block"),
+    MathDisplay("math_display");
+
+    /** The string value. */
+    private final String value;
+
+    BlockType(final String value) {
+        this.value = value;
+    }
+
+    /** Returns the string value. */
+    @JsonValue
+    public String getValue() {
+        return value;
+    }
+
+    /** Creates an instance from a string value. */
+    @JsonCreator
+    public static BlockType fromValue(final String value) {
+        for (BlockType e : values()) {
+            if (e.value.equalsIgnoreCase(value)) {
+                return e;
+            }
+        }
+        throw new IllegalArgumentException("Unknown value: " + value);
+    }
+
+    /** Returns the wire-format string value (matches JSON serialization). */
+    @Override
+    public String toString() {
+        return value;
+    }
+}

packages/java/dev/kreuzberg/BoundingBox.java

@@ -0,0 +1,88 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * Bounding box coordinates for element positioning.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = BoundingBox.Builder.class)
+public record BoundingBox(
+    /**
+     * Left x-coordinate
+     */
+    @JsonProperty("x0") double x0,
+    /**
+     * Bottom y-coordinate
+     */
+    @JsonProperty("y0") double y0,
+    /**
+     * Right x-coordinate
+     */
+    @JsonProperty("x1") double x1,
+    /**
+     * Top y-coordinate
+     */
+    @JsonProperty("y1") double y1
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private double x0 = 0.0;
+private double y0 = 0.0;
+private double x1 = 0.0;
+private double y1 = 0.0;
+
+        /** Sets the x0 field. */
+        @JsonProperty("x0")
+        public Builder withX0(final double value) {
+            this.x0 = value;
+            return this;
+        }
+
+        /** Sets the y0 field. */
+        @JsonProperty("y0")
+        public Builder withY0(final double value) {
+            this.y0 = value;
+            return this;
+        }
+
+        /** Sets the x1 field. */
+        @JsonProperty("x1")
+        public Builder withX1(final double value) {
+            this.x1 = value;
+            return this;
+        }
+
+        /** Sets the y1 field. */
+        @JsonProperty("y1")
+        public Builder withY1(final double value) {
+            this.y1 = value;
+            return this;
+        }
+
+        /** Builds the BoundingBox instance. */
+        public BoundingBox build() {
+            return new BoundingBox(
+                x0,
+                y0,
+                x1,
+                y1
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/ByteArrayToIntArraySerializer.java

@@ -0,0 +1,35 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.core.JsonGenerator;
+import com.fasterxml.jackson.databind.SerializerProvider;
+import com.fasterxml.jackson.databind.ser.std.StdSerializer;
+
+/**
+ * Serialises {@code byte[]} as a JSON array of integers.
+ *
+ * <p>Jackson's default serialiser encodes {@code byte[]} as a base64 string, but
+ * Rust's {@code serde} for {@code Vec<u8>} expects {@code [72, 101, 108, ...]}.
+ * Annotate any {@code byte[]} field sent to the FFI layer with
+ * {@code @JsonSerialize(using = ByteArrayToIntArraySerializer.class)}.
+ */
+public class ByteArrayToIntArraySerializer extends StdSerializer<byte[]> {
+    /** Default constructor required by Jackson. */
+    public ByteArrayToIntArraySerializer() {
+        super(byte[].class);
+    }
+
+    @Override
+    public void serialize(final byte[] value, final JsonGenerator gen,
+            final SerializerProvider provider) throws java.io.IOException {
+        gen.writeStartArray();
+        for (byte b : value) {
+            gen.writeNumber(b & 0xFF);
+        }
+        gen.writeEndArray();
+    }
+}

packages/java/dev/kreuzberg/CacheException.java

@@ -0,0 +1,15 @@
+// DO NOT EDIT - auto-generated by alef
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+package dev.kreuzberg;
+
+public class CacheException extends KreuzbergErrorException {
+    /** Creates a new CacheException with the given message. */
+    public CacheException(final String message) {
+        super(message);
+    }
+
+    /** Creates a new CacheException with the given message and cause. */
+    public CacheException(final String message, final Throwable cause) {
+        super(message, cause);
+    }
+}

packages/java/dev/kreuzberg/CacheStats.java

@@ -0,0 +1,88 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = CacheStats.Builder.class)
+public record CacheStats(
+    @JsonProperty("total_files") long totalFiles,
+    @JsonProperty("total_size_mb") double totalSizeMb,
+    @JsonProperty("available_space_mb") double availableSpaceMb,
+    @JsonProperty("oldest_file_age_days") double oldestFileAgeDays,
+    @JsonProperty("newest_file_age_days") double newestFileAgeDays
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("total_files")
+private long totalFiles = 0;
+        @JsonProperty("total_size_mb")
+private double totalSizeMb = 0.0;
+        @JsonProperty("available_space_mb")
+private double availableSpaceMb = 0.0;
+        @JsonProperty("oldest_file_age_days")
+private double oldestFileAgeDays = 0.0;
+        @JsonProperty("newest_file_age_days")
+private double newestFileAgeDays = 0.0;
+
+        /** Sets the totalFiles field. */
+        @JsonProperty("total_files")
+        public Builder withTotalFiles(final long value) {
+            this.totalFiles = value;
+            return this;
+        }
+
+        /** Sets the totalSizeMb field. */
+        @JsonProperty("total_size_mb")
+        public Builder withTotalSizeMb(final double value) {
+            this.totalSizeMb = value;
+            return this;
+        }
+
+        /** Sets the availableSpaceMb field. */
+        @JsonProperty("available_space_mb")
+        public Builder withAvailableSpaceMb(final double value) {
+            this.availableSpaceMb = value;
+            return this;
+        }
+
+        /** Sets the oldestFileAgeDays field. */
+        @JsonProperty("oldest_file_age_days")
+        public Builder withOldestFileAgeDays(final double value) {
+            this.oldestFileAgeDays = value;
+            return this;
+        }
+
+        /** Sets the newestFileAgeDays field. */
+        @JsonProperty("newest_file_age_days")
+        public Builder withNewestFileAgeDays(final double value) {
+            this.newestFileAgeDays = value;
+            return this;
+        }
+
+        /** Builds the CacheStats instance. */
+        public CacheStats build() {
+            return new CacheStats(
+                totalFiles,
+                totalSizeMb,
+                availableSpaceMb,
+                oldestFileAgeDays,
+                newestFileAgeDays
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/CancelledException.java

@@ -0,0 +1,15 @@
+// DO NOT EDIT - auto-generated by alef
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+package dev.kreuzberg;
+
+public class CancelledException extends KreuzbergErrorException {
+    /** Creates a new CancelledException with the given message. */
+    public CancelledException(final String message) {
+        super(message);
+    }
+
+    /** Creates a new CancelledException with the given message and cause. */
+    public CancelledException(final String message, final Throwable cause) {
+        super(message, cause);
+    }
+}

packages/java/dev/kreuzberg/CellChange.java

@@ -0,0 +1,92 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * A single changed cell within a table.
+ *
+ * Defined here (rather than only in {@code crate.diff}) so {@code RevisionDelta} can
+ * reference it unconditionally, without requiring the {@code diff} Cargo feature.
+ * {@code crate.diff} re-exports this type verbatim.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = CellChange.Builder.class)
+public record CellChange(
+    /**
+     * Zero-based row index.
+     */
+    @JsonProperty("row") long row,
+    /**
+     * Zero-based column index.
+     */
+    @JsonProperty("col") long col,
+    /**
+     * Value before the change.
+     */
+    @JsonProperty("from") String from,
+    /**
+     * Value after the change.
+     */
+    @JsonProperty("to") String to
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private long row = 0;
+private long col = 0;
+private String from = "";
+private String to = "";
+
+        /** Sets the row field. */
+        @JsonProperty("row")
+        public Builder withRow(final long value) {
+            this.row = value;
+            return this;
+        }
+
+        /** Sets the col field. */
+        @JsonProperty("col")
+        public Builder withCol(final long value) {
+            this.col = value;
+            return this;
+        }
+
+        /** Sets the from field. */
+        @JsonProperty("from")
+        public Builder withFrom(final String value) {
+            this.from = value;
+            return this;
+        }
+
+        /** Sets the to field. */
+        @JsonProperty("to")
+        public Builder withTo(final String value) {
+            this.to = value;
+            return this;
+        }
+
+        /** Builds the CellChange instance. */
+        public CellChange build() {
+            return new CellChange(
+                row,
+                col,
+                from,
+                to
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/Chunk.java

@@ -0,0 +1,101 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * A text chunk with optional embedding and metadata.
+ *
+ * Chunks are created when chunking is enabled in {@code ExtractionConfig}. Each chunk
+ * contains the text content, optional embedding vector (if embedding generation
+ * is configured), and metadata about its position in the document.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = Chunk.Builder.class)
+public record Chunk(
+    /**
+     * The text content of this chunk.
+     */
+    @JsonProperty("content") String content,
+    /**
+     * Semantic structural classification of this chunk.
+     *
+     * Assigned by the heuristic classifier based on content patterns and
+     * heading context. Defaults to {@code ChunkType.Unknown} when no rule matches.
+     */
+    @Nullable @JsonProperty("chunk_type") ChunkType chunkType,
+    /**
+     * Optional embedding vector for this chunk.
+     *
+     * Only populated when {@code EmbeddingConfig} is provided in chunking configuration.
+     * The dimensionality depends on the chosen embedding model.
+     */
+    @Nullable @JsonProperty("embedding") List<Float> embedding,
+    /**
+     * Metadata about this chunk's position and properties.
+     */
+    @JsonProperty("metadata") ChunkMetadata metadata
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String content = "";
+        @JsonProperty("chunk_type")
+        @Nullable private ChunkType chunkType = ChunkType.Unknown;
+private List<Float> embedding = null;
+private ChunkMetadata metadata = null;
+
+        /** Sets the content field. */
+        @JsonProperty("content")
+        public Builder withContent(final String value) {
+            this.content = value;
+            return this;
+        }
+
+        /** Sets the chunkType field. */
+        @JsonProperty("chunk_type")
+        public Builder withChunkType(final @Nullable ChunkType value) {
+            this.chunkType = value;
+            return this;
+        }
+
+        /** Sets the embedding field. */
+        @JsonProperty("embedding")
+        public Builder withEmbedding(final @Nullable List<Float> value) {
+            this.embedding = value;
+            return this;
+        }
+
+        /** Sets the metadata field. */
+        @JsonProperty("metadata")
+        public Builder withMetadata(final ChunkMetadata value) {
+            this.metadata = value;
+            return this;
+        }
+
+        /** Builds the Chunk instance. */
+        public Chunk build() {
+            return new Chunk(
+                content,
+                chunkType,
+                embedding,
+                metadata
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/ChunkMetadata.java

@@ -0,0 +1,177 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Metadata about a chunk's position in the original document.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ChunkMetadata.Builder.class)
+public record ChunkMetadata(
+    /**
+     * Byte offset where this chunk starts in the original text (UTF-8 valid boundary).
+     */
+    @JsonProperty("byte_start") long byteStart,
+    /**
+     * Byte offset where this chunk ends in the original text (UTF-8 valid boundary).
+     */
+    @JsonProperty("byte_end") long byteEnd,
+    /**
+     * Number of tokens in this chunk (if available).
+     *
+     * This is calculated by the embedding model's tokenizer if embeddings are enabled.
+     */
+    @Nullable @JsonProperty("token_count") Long tokenCount,
+    /**
+     * Zero-based index of this chunk in the document.
+     */
+    @JsonProperty("chunk_index") long chunkIndex,
+    /**
+     * Total number of chunks in the document.
+     */
+    @JsonProperty("total_chunks") long totalChunks,
+    /**
+     * First page number this chunk spans (1-indexed).
+     *
+     * Only populated when page tracking is enabled in extraction configuration.
+     */
+    @Nullable @JsonProperty("first_page") Integer firstPage,
+    /**
+     * Last page number this chunk spans (1-indexed, equal to first_page for single-page chunks).
+     *
+     * Only populated when page tracking is enabled in extraction configuration.
+     */
+    @Nullable @JsonProperty("last_page") Integer lastPage,
+    /**
+     * Heading context when using Markdown chunker.
+     *
+     * Contains the heading hierarchy this chunk falls under.
+     * Only populated when {@code ChunkerType.Markdown} is used.
+     */
+    @Nullable @JsonProperty("heading_context") HeadingContext headingContext,
+    /**
+     * Indices into {@code ExtractionResult.images} for images on pages covered by this chunk.
+     *
+     * Contains zero-based indices into the top-level {@code images} collection for every
+     * image whose {@code page_number} falls within {@code [first_page, last_page]}.
+     * Empty when image extraction is disabled or the chunk spans no pages with images.
+     */
+    @Nullable @JsonProperty("image_indices") List<Integer> imageIndices
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("byte_start")
+private long byteStart = 0;
+        @JsonProperty("byte_end")
+private long byteEnd = 0;
+        @JsonProperty("token_count")
+private Long tokenCount = null;
+        @JsonProperty("chunk_index")
+private long chunkIndex = 0;
+        @JsonProperty("total_chunks")
+private long totalChunks = 0;
+        @JsonProperty("first_page")
+private Integer firstPage = null;
+        @JsonProperty("last_page")
+private Integer lastPage = null;
+        @JsonProperty("heading_context")
+        @Nullable private HeadingContext headingContext = null;
+        @JsonProperty("image_indices")
+private List<Integer> imageIndices = null;
+
+        /** Sets the byteStart field. */
+        @JsonProperty("byte_start")
+        public Builder withByteStart(final long value) {
+            this.byteStart = value;
+            return this;
+        }
+
+        /** Sets the byteEnd field. */
+        @JsonProperty("byte_end")
+        public Builder withByteEnd(final long value) {
+            this.byteEnd = value;
+            return this;
+        }
+
+        /** Sets the tokenCount field. */
+        @JsonProperty("token_count")
+        public Builder withTokenCount(final @Nullable long value) {
+            this.tokenCount = value;
+            return this;
+        }
+
+        /** Sets the chunkIndex field. */
+        @JsonProperty("chunk_index")
+        public Builder withChunkIndex(final long value) {
+            this.chunkIndex = value;
+            return this;
+        }
+
+        /** Sets the totalChunks field. */
+        @JsonProperty("total_chunks")
+        public Builder withTotalChunks(final long value) {
+            this.totalChunks = value;
+            return this;
+        }
+
+        /** Sets the firstPage field. */
+        @JsonProperty("first_page")
+        public Builder withFirstPage(final @Nullable int value) {
+            this.firstPage = value;
+            return this;
+        }
+
+        /** Sets the lastPage field. */
+        @JsonProperty("last_page")
+        public Builder withLastPage(final @Nullable int value) {
+            this.lastPage = value;
+            return this;
+        }
+
+        /** Sets the headingContext field. */
+        @JsonProperty("heading_context")
+        public Builder withHeadingContext(final @Nullable HeadingContext value) {
+            this.headingContext = value;
+            return this;
+        }
+
+        /** Sets the imageIndices field. */
+        @JsonProperty("image_indices")
+        public Builder withImageIndices(final @Nullable List<Integer> value) {
+            this.imageIndices = value;
+            return this;
+        }
+
+        /** Builds the ChunkMetadata instance. */
+        public ChunkMetadata build() {
+            return new ChunkMetadata(
+                byteStart,
+                byteEnd,
+                tokenCount,
+                chunkIndex,
+                totalChunks,
+                firstPage,
+                lastPage,
+                headingContext,
+                imageIndices
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/ChunkSizing.java

@@ -0,0 +1,41 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonSubTypes;
+import com.fasterxml.jackson.annotation.JsonTypeInfo;
+import java.util.Optional;
+
+/**
+ * How chunk size is measured.
+ *
+ * Defaults to {@code Characters} (Unicode character count). When using token-based sizing,
+ * chunks are sized by token count according to the specified tokenizer.
+ *
+ * Token-based sizing uses HuggingFace tokenizers loaded at runtime. Any tokenizer
+ * available on HuggingFace Hub can be used, including OpenAI-compatible tokenizers
+ * (e.g., {@code Xenova/gpt-4o}, {@code Xenova/cl100k_base}).
+ */
+@JsonTypeInfo(use = JsonTypeInfo.Id.NAME, property = "type", visible = false)
+@JsonSubTypes({
+    @JsonSubTypes.Type(value = ChunkSizing.Characters.class, name = "characters"),
+    @JsonSubTypes.Type(value = ChunkSizing.Tokenizer.class, name = "tokenizer")
+})
+@com.fasterxml.jackson.annotation.JsonIgnoreProperties(ignoreUnknown = true)
+public sealed interface ChunkSizing {
+
+    /** Size measured in Unicode characters (default). */
+    record Characters() implements ChunkSizing {
+    }
+
+    /** Size measured in tokens from a HuggingFace tokenizer. */
+    record Tokenizer(
+        @JsonProperty("model") String model,
+        @JsonProperty("cache_dir") Optional<java.nio.file.Path> cacheDir
+    ) implements ChunkSizing {
+    }
+}

packages/java/dev/kreuzberg/ChunkType.java

@@ -0,0 +1,75 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonCreator;
+import com.fasterxml.jackson.annotation.JsonValue;
+
+/**
+ * Semantic structural classification of a text chunk.
+ *
+ * Assigned by the heuristic classifier in {@code chunking.classifier}.
+ * Defaults to {@code Unknown} when no rule matches.
+ * Designed to be extended in future versions without breaking changes.
+ */
+public enum ChunkType {
+    /** Section heading or document title. */
+    Heading("heading"),
+    /** Party list: names, addresses, and signatories. */
+    PartyList("party_list"),
+    /** Definition clause ("X means…", "X shall mean…"). */
+    Definitions("definitions"),
+    /** Operative clause containing legal/contractual action verbs. */
+    OperativeClause("operative_clause"),
+    /** Signature block with signatures, names, and dates. */
+    SignatureBlock("signature_block"),
+    /** Schedule, annex, appendix, or exhibit section. */
+    Schedule("schedule"),
+    /** Table-like content with aligned columns or repeated patterns. */
+    TableLike("table_like"),
+    /** Mathematical formula or equation. */
+    Formula("formula"),
+    /** Code block or preformatted content. */
+    CodeBlock("code_block"),
+    /** Embedded or referenced image content. */
+    Image("image"),
+    /** Organizational chart or hierarchy diagram. */
+    OrgChart("org_chart"),
+    /** Diagram, figure, or visual illustration. */
+    Diagram("diagram"),
+    /** Unclassified or mixed content. */
+    Unknown("unknown");
+
+    /** The string value. */
+    private final String value;
+
+    ChunkType(final String value) {
+        this.value = value;
+    }
+
+    /** Returns the string value. */
+    @JsonValue
+    public String getValue() {
+        return value;
+    }
+
+    /** Creates an instance from a string value. */
+    @JsonCreator
+    public static ChunkType fromValue(final String value) {
+        for (ChunkType e : values()) {
+            if (e.value.equalsIgnoreCase(value)) {
+                return e;
+            }
+        }
+        throw new IllegalArgumentException("Unknown value: " + value);
+    }
+
+    /** Returns the wire-format string value (matches JSON serialization). */
+    @Override
+    public String toString() {
+        return value;
+    }
+}

packages/java/dev/kreuzberg/ChunkerType.java

@@ -0,0 +1,62 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonCreator;
+import com.fasterxml.jackson.annotation.JsonValue;
+
+/**
+ * Type of text chunker to use.
+ *
+ * # Variants
+ *
+ * * {@code Text} - Generic text splitter, splits on whitespace and punctuation
+ * * {@code Markdown} - Markdown-aware splitter, preserves formatting and structure
+ * * {@code Yaml} - YAML-aware splitter, creates one chunk per top-level key
+ * * {@code Semantic} - Topic-aware chunker. With an {@code EmbeddingConfig}, splits at
+ *   embedding-based topic shifts tuned by {@code topic_threshold} (default 0.75,
+ *   lower = more splits). Without an embedding, falls back to a
+ *   structural-boundary heuristic (ALL-CAPS headers, numbered sections,
+ *   blank-line paragraphs) and merges groups into chunks capped at
+ *   {@code max_characters} (default 1000). {@code topic_threshold} has no effect in the
+ *   fallback path. For best results, pair with an embedding model.
+ */
+public enum ChunkerType {
+    Text("text"),
+    Markdown("markdown"),
+    Yaml("yaml"),
+    Semantic("semantic");
+
+    /** The string value. */
+    private final String value;
+
+    ChunkerType(final String value) {
+        this.value = value;
+    }
+
+    /** Returns the string value. */
+    @JsonValue
+    public String getValue() {
+        return value;
+    }
+
+    /** Creates an instance from a string value. */
+    @JsonCreator
+    public static ChunkerType fromValue(final String value) {
+        for (ChunkerType e : values()) {
+            if (e.value.equalsIgnoreCase(value)) {
+                return e;
+            }
+        }
+        throw new IllegalArgumentException("Unknown value: " + value);
+    }
+
+    /** Returns the wire-format string value (matches JSON serialization). */
+    @Override
+    public String toString() {
+        return value;
+    }
+}

packages/java/dev/kreuzberg/ChunkingConfig.java

@@ -0,0 +1,203 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Chunking configuration.
+ *
+ * Configures text chunking for document content, including chunk size,
+ * overlap, trimming behavior, and optional embeddings.
+ *
+ * Use {@code ..Default.default()} when constructing to allow for future field additions:
+ *
+ * let config = ChunkingConfig {
+ *     max_characters: 500,
+ *     ..Default.default()
+ * };
+ *
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ChunkingConfig.Builder.class)
+public record ChunkingConfig(
+    /**
+     * Maximum size per chunk (in units determined by {@code sizing}).
+     *
+     * When {@code sizing} is {@code Characters} (default), this is the max character count.
+     * When using token-based sizing, this is the max token count.
+     *
+     * Default: 1000
+     */
+    @Nullable @JsonProperty("max_chars") Long maxCharacters,
+    /**
+     * Overlap between chunks (in units determined by {@code sizing}).
+     *
+     * Default: 200
+     */
+    @Nullable @JsonProperty("max_overlap") Long overlap,
+    /**
+     * Whether to trim whitespace from chunk boundaries.
+     *
+     * Default: true
+     */
+    @Nullable @JsonProperty("trim") Boolean trim,
+    /**
+     * Type of chunker to use (Text or Markdown).
+     *
+     * Default: Text
+     */
+    @Nullable @JsonProperty("chunker_type") ChunkerType chunkerType,
+    /**
+     * Optional embedding configuration for chunk embeddings.
+     */
+    @Nullable @JsonProperty("embedding") EmbeddingConfig embedding,
+    /**
+     * Use a preset configuration (overrides individual settings if provided).
+     */
+    @Nullable @JsonProperty("preset") String preset,
+    /**
+     * How to measure chunk size.
+     *
+     * Default: {@code Characters} (Unicode character count).
+     * Enable {@code chunking-tiktoken} or {@code chunking-tokenizers} features for token-based sizing.
+     */
+    @Nullable @JsonProperty("sizing") ChunkSizing sizing,
+    /**
+     * When {@code true} and {@code chunker_type} is {@code Markdown}, prepend the heading hierarchy
+     * path (e.g. {@code "# Title > ## Section\n\n"}) to each chunk's content string.
+     *
+     * This is useful for RAG pipelines where each chunk needs self-contained
+     * context about its position in the document structure.
+     *
+     * Default: {@code false}
+     */
+    @Nullable @JsonProperty("prepend_heading_context") Boolean prependHeadingContext,
+    /**
+     * Optional cosine similarity threshold for semantic topic boundary detection.
+     *
+     * Only used when {@code chunker_type} is {@code Semantic} and an {@code EmbeddingConfig} is
+     * provided. You almost never need to set this. When omitted, defaults to
+     * {@code 0.75} which works well for most documents. Lower values detect more
+     * topic boundaries (more, smaller chunks); higher values detect fewer.
+     * Range: {@code 0.0..=1.0}.
+     */
+    @Nullable @JsonProperty("topic_threshold") Float topicThreshold
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+    public ChunkingConfig{
+        if (maxCharacters == null) maxCharacters = 1000L;
+        if (overlap == null) overlap = 200L;
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("max_chars")
+private Long maxCharacters = null;
+        @JsonProperty("max_overlap")
+private Long overlap = null;
+private Boolean trim = null;
+        @JsonProperty("chunker_type")
+        @Nullable private ChunkerType chunkerType = ChunkerType.Text;
+private EmbeddingConfig embedding = null;
+private String preset = null;
+        @Nullable private ChunkSizing sizing = new ChunkSizing.Characters();
+        @JsonProperty("prepend_heading_context")
+private Boolean prependHeadingContext = null;
+        @JsonProperty("topic_threshold")
+private Float topicThreshold = null;
+
+        /** Sets the maxCharacters field. */
+        @JsonProperty("max_chars")
+        public Builder withMaxCharacters(final @Nullable Long value) {
+            this.maxCharacters = value;
+            return this;
+        }
+
+        /** Sets the overlap field. */
+        @JsonProperty("max_overlap")
+        public Builder withOverlap(final @Nullable Long value) {
+            this.overlap = value;
+            return this;
+        }
+
+        /** Sets the trim field. */
+        @JsonProperty("trim")
+        public Builder withTrim(final @Nullable Boolean value) {
+            this.trim = value;
+            return this;
+        }
+
+        /** Sets the chunkerType field. */
+        @JsonProperty("chunker_type")
+        public Builder withChunkerType(final @Nullable ChunkerType value) {
+            this.chunkerType = value;
+            return this;
+        }
+
+        /** Sets the embedding field. */
+        @JsonProperty("embedding")
+        public Builder withEmbedding(final @Nullable EmbeddingConfig value) {
+            this.embedding = value;
+            return this;
+        }
+
+        /** Sets the preset field. */
+        @JsonProperty("preset")
+        public Builder withPreset(final @Nullable String value) {
+            this.preset = value;
+            return this;
+        }
+
+        /** Sets the sizing field. */
+        @JsonProperty("sizing")
+        public Builder withSizing(final @Nullable ChunkSizing value) {
+            this.sizing = value;
+            return this;
+        }
+
+        /** Sets the prependHeadingContext field. */
+        @JsonProperty("prepend_heading_context")
+        public Builder withPrependHeadingContext(final @Nullable Boolean value) {
+            this.prependHeadingContext = value;
+            return this;
+        }
+
+        /** Sets the topicThreshold field. */
+        @JsonProperty("topic_threshold")
+        public Builder withTopicThreshold(final @Nullable Float value) {
+            this.topicThreshold = value;
+            return this;
+        }
+
+        /** Builds the ChunkingConfig instance. */
+        public ChunkingConfig build() {
+            return new ChunkingConfig(
+                maxCharacters,
+                overlap,
+                trim,
+                chunkerType,
+                embedding,
+                preset,
+                sizing,
+                prependHeadingContext,
+                topicThreshold
+            );
+        }
+    }
+    // CPD-ON
+    public static ChunkingConfig defaultInstance() {
+        throw new UnsupportedOperationException("defaultInstance is not yet bridged via JNI; use the Builder instead.");
+    }
+}

packages/java/dev/kreuzberg/CitationMetadata.java

@@ -0,0 +1,100 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Citation file metadata (RIS, PubMed, EndNote).
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = CitationMetadata.Builder.class)
+public record CitationMetadata(
+    @JsonProperty("citation_count") long citationCount,
+    @Nullable @JsonProperty("format") String format,
+    @Nullable @JsonProperty("authors") List<String> authors,
+    @Nullable @JsonProperty("year_range") YearRange yearRange,
+    @Nullable @JsonProperty("dois") List<String> dois,
+    @Nullable @JsonProperty("keywords") List<String> keywords
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("citation_count")
+private long citationCount = 0;
+private String format = null;
+private List<String> authors = null;
+        @JsonProperty("year_range")
+private YearRange yearRange = null;
+private List<String> dois = null;
+private List<String> keywords = null;
+
+        /** Sets the citationCount field. */
+        @JsonProperty("citation_count")
+        public Builder withCitationCount(final long value) {
+            this.citationCount = value;
+            return this;
+        }
+
+        /** Sets the format field. */
+        @JsonProperty("format")
+        public Builder withFormat(final @Nullable String value) {
+            this.format = value;
+            return this;
+        }
+
+        /** Sets the authors field. */
+        @JsonProperty("authors")
+        public Builder withAuthors(final @Nullable List<String> value) {
+            this.authors = value;
+            return this;
+        }
+
+        /** Sets the yearRange field. */
+        @JsonProperty("year_range")
+        public Builder withYearRange(final @Nullable YearRange value) {
+            this.yearRange = value;
+            return this;
+        }
+
+        /** Sets the dois field. */
+        @JsonProperty("dois")
+        public Builder withDois(final @Nullable List<String> value) {
+            this.dois = value;
+            return this;
+        }
+
+        /** Sets the keywords field. */
+        @JsonProperty("keywords")
+        public Builder withKeywords(final @Nullable List<String> value) {
+            this.keywords = value;
+            return this;
+        }
+
+        /** Builds the CitationMetadata instance. */
+        public CitationMetadata build() {
+            return new CitationMetadata(
+                citationCount,
+                format,
+                authors,
+                yearRange,
+                dois,
+                keywords
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/CodeContentMode.java

@@ -0,0 +1,54 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonCreator;
+import com.fasterxml.jackson.annotation.JsonValue;
+
+/**
+ * Content rendering mode for code extraction.
+ *
+ * Controls how extracted code content is represented in the {@code content} field
+ * of {@code ExtractionResult}.
+ */
+public enum CodeContentMode {
+    /** Use TSLP semantic chunks as content (default). */
+    Chunks("chunks"),
+    /** Use raw source code as content. */
+    Raw("raw"),
+    /** Emit function/class headings + docstrings (no code bodies). */
+    Structure("structure");
+
+    /** The string value. */
+    private final String value;
+
+    CodeContentMode(final String value) {
+        this.value = value;
+    }
+
+    /** Returns the string value. */
+    @JsonValue
+    public String getValue() {
+        return value;
+    }
+
+    /** Creates an instance from a string value. */
+    @JsonCreator
+    public static CodeContentMode fromValue(final String value) {
+        for (CodeContentMode e : values()) {
+            if (e.value.equalsIgnoreCase(value)) {
+                return e;
+            }
+        }
+        throw new IllegalArgumentException("Unknown value: " + value);
+    }
+
+    /** Returns the wire-format string value (matches JSON serialization). */
+    @Override
+    public String toString() {
+        return value;
+    }
+}

packages/java/dev/kreuzberg/ContentFilterConfig.java

@@ -0,0 +1,138 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Cross-extractor content filtering configuration.
+ *
+ * Controls whether "furniture" content (headers, footers, page numbers,
+ * watermarks, repeating text) is included in or stripped from extraction
+ * results. Applies across all extractors (PDF, DOCX, RTF, ODT, HTML, etc.)
+ * with format-specific implementation.
+ *
+ * When {@code None} on {@code ExtractionConfig}, each extractor uses its current
+ * default behavior unchanged.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ContentFilterConfig.Builder.class)
+public record ContentFilterConfig(
+    /**
+     * Include running headers in extraction output.
+     *
+     * - PDF: Disables top-margin furniture stripping and prevents the layout
+     *   model from treating {@code PageHeader}-classified regions as furniture.
+     * - DOCX: Includes document headers in text output.
+     * - RTF/ODT: Headers already included; this is a no-op when true.
+     * - HTML/EPUB: Keeps {@code <header>} element content.
+     *
+     * Default: {@code false} (headers are stripped or excluded).
+     */
+    @Nullable @JsonProperty("include_headers") Boolean includeHeaders,
+    /**
+     * Include running footers in extraction output.
+     *
+     * - PDF: Disables bottom-margin furniture stripping and prevents the layout
+     *   model from treating {@code PageFooter}-classified regions as furniture.
+     * - DOCX: Includes document footers in text output.
+     * - RTF/ODT: Footers already included; this is a no-op when true.
+     * - HTML/EPUB: Keeps {@code <footer>} element content.
+     *
+     * Default: {@code false} (footers are stripped or excluded).
+     */
+    @Nullable @JsonProperty("include_footers") Boolean includeFooters,
+    /**
+     * Enable the heuristic cross-page repeating text detector.
+     *
+     * When {@code true} (default), text that repeats verbatim across a supermajority
+     * of pages is classified as furniture and stripped.  Disable this if brand
+     * names or repeated headings are being incorrectly removed by the heuristic.
+     *
+     * Note: when a layout-detection model is active, the model may independently
+     * classify page-header / page-footer regions as furniture on a per-page basis.
+     * To preserve those regions, set {@code include_headers = true}, {@code include_footers = true},
+     * or both, in addition to disabling this flag.
+     *
+     * Primarily affects PDF extraction.
+     *
+     * Default: {@code true}.
+     */
+    @Nullable @JsonProperty("strip_repeating_text") Boolean stripRepeatingText,
+    /**
+     * Include watermark text in extraction output.
+     *
+     * - PDF: Keeps watermark artifacts and arXiv identifiers.
+     * - Other formats: No effect currently.
+     *
+     * Default: {@code false} (watermarks are stripped).
+     */
+    @Nullable @JsonProperty("include_watermarks") Boolean includeWatermarks
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("include_headers")
+private Boolean includeHeaders = null;
+        @JsonProperty("include_footers")
+private Boolean includeFooters = null;
+        @JsonProperty("strip_repeating_text")
+private Boolean stripRepeatingText = null;
+        @JsonProperty("include_watermarks")
+private Boolean includeWatermarks = null;
+
+        /** Sets the includeHeaders field. */
+        @JsonProperty("include_headers")
+        public Builder withIncludeHeaders(final @Nullable Boolean value) {
+            this.includeHeaders = value;
+            return this;
+        }
+
+        /** Sets the includeFooters field. */
+        @JsonProperty("include_footers")
+        public Builder withIncludeFooters(final @Nullable Boolean value) {
+            this.includeFooters = value;
+            return this;
+        }
+
+        /** Sets the stripRepeatingText field. */
+        @JsonProperty("strip_repeating_text")
+        public Builder withStripRepeatingText(final @Nullable Boolean value) {
+            this.stripRepeatingText = value;
+            return this;
+        }
+
+        /** Sets the includeWatermarks field. */
+        @JsonProperty("include_watermarks")
+        public Builder withIncludeWatermarks(final @Nullable Boolean value) {
+            this.includeWatermarks = value;
+            return this;
+        }
+
+        /** Builds the ContentFilterConfig instance. */
+        public ContentFilterConfig build() {
+            return new ContentFilterConfig(
+                includeHeaders,
+                includeFooters,
+                stripRepeatingText,
+                includeWatermarks
+            );
+        }
+    }
+    // CPD-ON
+    public static ContentFilterConfig defaultInstance() {
+        throw new UnsupportedOperationException("defaultInstance is not yet bridged via JNI; use the Builder instead.");
+    }
+}

packages/java/dev/kreuzberg/ContentLayer.java

@@ -0,0 +1,55 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonCreator;
+import com.fasterxml.jackson.annotation.JsonValue;
+
+/**
+ * Content layer classification for document nodes.
+ *
+ * Replaces separate body/furniture arrays with per-node granularity.
+ */
+public enum ContentLayer {
+    /** Main document body content. */
+    Body("body"),
+    /** Page/section header (running header). */
+    Header("header"),
+    /** Page/section footer (running footer). */
+    Footer("footer"),
+    /** Footnote content. */
+    Footnote("footnote");
+
+    /** The string value. */
+    private final String value;
+
+    ContentLayer(final String value) {
+        this.value = value;
+    }
+
+    /** Returns the string value. */
+    @JsonValue
+    public String getValue() {
+        return value;
+    }
+
+    /** Creates an instance from a string value. */
+    @JsonCreator
+    public static ContentLayer fromValue(final String value) {
+        for (ContentLayer e : values()) {
+            if (e.value.equalsIgnoreCase(value)) {
+                return e;
+            }
+        }
+        throw new IllegalArgumentException("Unknown value: " + value);
+    }
+
+    /** Returns the wire-format string value (matches JSON serialization). */
+    @Override
+    public String toString() {
+        return value;
+    }
+}

packages/java/dev/kreuzberg/ContributorRole.java

@@ -0,0 +1,57 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * JATS contributor with role.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ContributorRole.Builder.class)
+public record ContributorRole(
+    @JsonProperty("name") String name,
+    @Nullable @JsonProperty("role") String role
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String name = "";
+private String role = null;
+
+        /** Sets the name field. */
+        @JsonProperty("name")
+        public Builder withName(final String value) {
+            this.name = value;
+            return this;
+        }
+
+        /** Sets the role field. */
+        @JsonProperty("role")
+        public Builder withRole(final @Nullable String value) {
+            this.role = value;
+            return this;
+        }
+
+        /** Builds the ContributorRole instance. */
+        public ContributorRole build() {
+            return new ContributorRole(
+                name,
+                role
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/ConversionErrorException.java

@@ -0,0 +1,20 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+/** Exception thrown when type conversion fails. */
+@SuppressWarnings("checkstyle:LineLength")
+public class ConversionErrorException extends KreuzbergRsException {
+    /** Creates a new ConversionErrorException. */
+    public ConversionErrorException(final String message) {
+        super(2, message);
+    }
+
+    /** Creates a new ConversionErrorException with a cause. */
+    public ConversionErrorException(final String message, final Throwable cause) {
+        super(message, cause);
+    }
+}

packages/java/dev/kreuzberg/CoreProperties.java

@@ -0,0 +1,238 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Dublin Core metadata from docProps/core.xml
+ *
+ * Contains standard metadata fields defined by the Dublin Core standard
+ * and Office-specific extensions.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = CoreProperties.Builder.class)
+public record CoreProperties(
+    /**
+     * Document title
+     */
+    @Nullable @JsonProperty("title") String title,
+    /**
+     * Document subject/topic
+     */
+    @Nullable @JsonProperty("subject") String subject,
+    /**
+     * Document creator/author
+     */
+    @Nullable @JsonProperty("creator") String creator,
+    /**
+     * Keywords or tags
+     */
+    @Nullable @JsonProperty("keywords") String keywords,
+    /**
+     * Document description/abstract
+     */
+    @Nullable @JsonProperty("description") String description,
+    /**
+     * User who last modified the document
+     */
+    @Nullable @JsonProperty("last_modified_by") String lastModifiedBy,
+    /**
+     * Revision number
+     */
+    @Nullable @JsonProperty("revision") String revision,
+    /**
+     * Creation timestamp (ISO 8601)
+     */
+    @Nullable @JsonProperty("created") String created,
+    /**
+     * Last modification timestamp (ISO 8601)
+     */
+    @Nullable @JsonProperty("modified") String modified,
+    /**
+     * Document category
+     */
+    @Nullable @JsonProperty("category") String category,
+    /**
+     * Content status (Draft, Final, etc.)
+     */
+    @Nullable @JsonProperty("content_status") String contentStatus,
+    /**
+     * Document language
+     */
+    @Nullable @JsonProperty("language") String language,
+    /**
+     * Unique identifier
+     */
+    @Nullable @JsonProperty("identifier") String identifier,
+    /**
+     * Document version
+     */
+    @Nullable @JsonProperty("version") String version,
+    /**
+     * Last print timestamp (ISO 8601)
+     */
+    @Nullable @JsonProperty("last_printed") String lastPrinted
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String title = null;
+private String subject = null;
+private String creator = null;
+private String keywords = null;
+private String description = null;
+        @JsonProperty("last_modified_by")
+private String lastModifiedBy = null;
+private String revision = null;
+private String created = null;
+private String modified = null;
+private String category = null;
+        @JsonProperty("content_status")
+private String contentStatus = null;
+private String language = null;
+private String identifier = null;
+private String version = null;
+        @JsonProperty("last_printed")
+private String lastPrinted = null;
+
+        /** Sets the title field. */
+        @JsonProperty("title")
+        public Builder withTitle(final @Nullable String value) {
+            this.title = value;
+            return this;
+        }
+
+        /** Sets the subject field. */
+        @JsonProperty("subject")
+        public Builder withSubject(final @Nullable String value) {
+            this.subject = value;
+            return this;
+        }
+
+        /** Sets the creator field. */
+        @JsonProperty("creator")
+        public Builder withCreator(final @Nullable String value) {
+            this.creator = value;
+            return this;
+        }
+
+        /** Sets the keywords field. */
+        @JsonProperty("keywords")
+        public Builder withKeywords(final @Nullable String value) {
+            this.keywords = value;
+            return this;
+        }
+
+        /** Sets the description field. */
+        @JsonProperty("description")
+        public Builder withDescription(final @Nullable String value) {
+            this.description = value;
+            return this;
+        }
+
+        /** Sets the lastModifiedBy field. */
+        @JsonProperty("last_modified_by")
+        public Builder withLastModifiedBy(final @Nullable String value) {
+            this.lastModifiedBy = value;
+            return this;
+        }
+
+        /** Sets the revision field. */
+        @JsonProperty("revision")
+        public Builder withRevision(final @Nullable String value) {
+            this.revision = value;
+            return this;
+        }
+
+        /** Sets the created field. */
+        @JsonProperty("created")
+        public Builder withCreated(final @Nullable String value) {
+            this.created = value;
+            return this;
+        }
+
+        /** Sets the modified field. */
+        @JsonProperty("modified")
+        public Builder withModified(final @Nullable String value) {
+            this.modified = value;
+            return this;
+        }
+
+        /** Sets the category field. */
+        @JsonProperty("category")
+        public Builder withCategory(final @Nullable String value) {
+            this.category = value;
+            return this;
+        }
+
+        /** Sets the contentStatus field. */
+        @JsonProperty("content_status")
+        public Builder withContentStatus(final @Nullable String value) {
+            this.contentStatus = value;
+            return this;
+        }
+
+        /** Sets the language field. */
+        @JsonProperty("language")
+        public Builder withLanguage(final @Nullable String value) {
+            this.language = value;
+            return this;
+        }
+
+        /** Sets the identifier field. */
+        @JsonProperty("identifier")
+        public Builder withIdentifier(final @Nullable String value) {
+            this.identifier = value;
+            return this;
+        }
+
+        /** Sets the version field. */
+        @JsonProperty("version")
+        public Builder withVersion(final @Nullable String value) {
+            this.version = value;
+            return this;
+        }
+
+        /** Sets the lastPrinted field. */
+        @JsonProperty("last_printed")
+        public Builder withLastPrinted(final @Nullable String value) {
+            this.lastPrinted = value;
+            return this;
+        }
+
+        /** Builds the CoreProperties instance. */
+        public CoreProperties build() {
+            return new CoreProperties(
+                title,
+                subject,
+                creator,
+                keywords,
+                description,
+                lastModifiedBy,
+                revision,
+                created,
+                modified,
+                category,
+                contentStatus,
+                language,
+                identifier,
+                version,
+                lastPrinted
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/CsvMetadata.java

@@ -0,0 +1,92 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * CSV/TSV file metadata.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = CsvMetadata.Builder.class)
+public record CsvMetadata(
+    @JsonProperty("row_count") int rowCount,
+    @JsonProperty("column_count") int columnCount,
+    @Nullable @JsonProperty("delimiter") String delimiter,
+    @JsonProperty("has_header") boolean hasHeader,
+    @Nullable @JsonProperty("column_types") List<String> columnTypes
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("row_count")
+private int rowCount = 0;
+        @JsonProperty("column_count")
+private int columnCount = 0;
+private String delimiter = null;
+        @JsonProperty("has_header")
+private boolean hasHeader = false;
+        @JsonProperty("column_types")
+private List<String> columnTypes = null;
+
+        /** Sets the rowCount field. */
+        @JsonProperty("row_count")
+        public Builder withRowCount(final int value) {
+            this.rowCount = value;
+            return this;
+        }
+
+        /** Sets the columnCount field. */
+        @JsonProperty("column_count")
+        public Builder withColumnCount(final int value) {
+            this.columnCount = value;
+            return this;
+        }
+
+        /** Sets the delimiter field. */
+        @JsonProperty("delimiter")
+        public Builder withDelimiter(final @Nullable String value) {
+            this.delimiter = value;
+            return this;
+        }
+
+        /** Sets the hasHeader field. */
+        @JsonProperty("has_header")
+        public Builder withHasHeader(final boolean value) {
+            this.hasHeader = value;
+            return this;
+        }
+
+        /** Sets the columnTypes field. */
+        @JsonProperty("column_types")
+        public Builder withColumnTypes(final @Nullable List<String> value) {
+            this.columnTypes = value;
+            return this;
+        }
+
+        /** Builds the CsvMetadata instance. */
+        public CsvMetadata build() {
+            return new CsvMetadata(
+                rowCount,
+                columnCount,
+                delimiter,
+                hasHeader,
+                columnTypes
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/DbfFieldInfo.java

@@ -0,0 +1,57 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * dBASE field information.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = DbfFieldInfo.Builder.class)
+public record DbfFieldInfo(
+    @JsonProperty("name") String name,
+    @JsonProperty("field_type") String fieldType
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String name = "";
+        @JsonProperty("field_type")
+private String fieldType = "";
+
+        /** Sets the name field. */
+        @JsonProperty("name")
+        public Builder withName(final String value) {
+            this.name = value;
+            return this;
+        }
+
+        /** Sets the fieldType field. */
+        @JsonProperty("field_type")
+        public Builder withFieldType(final String value) {
+            this.fieldType = value;
+            return this;
+        }
+
+        /** Builds the DbfFieldInfo instance. */
+        public DbfFieldInfo build() {
+            return new DbfFieldInfo(
+                name,
+                fieldType
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/DbfMetadata.java

@@ -0,0 +1,70 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * dBASE (DBF) file metadata.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = DbfMetadata.Builder.class)
+public record DbfMetadata(
+    @JsonProperty("record_count") long recordCount,
+    @JsonProperty("field_count") long fieldCount,
+    @Nullable @JsonProperty("fields") List<DbfFieldInfo> fields
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("record_count")
+private long recordCount = 0;
+        @JsonProperty("field_count")
+private long fieldCount = 0;
+private List<DbfFieldInfo> fields = null;
+
+        /** Sets the recordCount field. */
+        @JsonProperty("record_count")
+        public Builder withRecordCount(final long value) {
+            this.recordCount = value;
+            return this;
+        }
+
+        /** Sets the fieldCount field. */
+        @JsonProperty("field_count")
+        public Builder withFieldCount(final long value) {
+            this.fieldCount = value;
+            return this;
+        }
+
+        /** Sets the fields field. */
+        @JsonProperty("fields")
+        public Builder withFields(final @Nullable List<DbfFieldInfo> value) {
+            this.fields = value;
+            return this;
+        }
+
+        /** Builds the DbfMetadata instance. */
+        public DbfMetadata build() {
+            return new DbfMetadata(
+                recordCount,
+                fieldCount,
+                fields
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/DetectResponse.java

@@ -0,0 +1,64 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * MIME type detection response.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = DetectResponse.Builder.class)
+public record DetectResponse(
+    /**
+     * Detected MIME type
+     */
+    @JsonProperty("mime_type") String mimeType,
+    /**
+     * Original filename (if provided)
+     */
+    @Nullable @JsonProperty("filename") String filename
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("mime_type")
+private String mimeType = "";
+private String filename = null;
+
+        /** Sets the mimeType field. */
+        @JsonProperty("mime_type")
+        public Builder withMimeType(final String value) {
+            this.mimeType = value;
+            return this;
+        }
+
+        /** Sets the filename field. */
+        @JsonProperty("filename")
+        public Builder withFilename(final @Nullable String value) {
+            this.filename = value;
+            return this;
+        }
+
+        /** Builds the DetectResponse instance. */
+        public DetectResponse build() {
+            return new DetectResponse(
+                mimeType,
+                filename
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/DetectionResult.java

@@ -0,0 +1,69 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * Page-level detection result containing all detections and page metadata.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = DetectionResult.Builder.class)
+public record DetectionResult(
+    @JsonProperty("page_width") int pageWidth,
+    @JsonProperty("page_height") int pageHeight,
+    @JsonProperty("detections") List<LayoutDetection> detections
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("page_width")
+private int pageWidth = 0;
+        @JsonProperty("page_height")
+private int pageHeight = 0;
+private List<LayoutDetection> detections = List.of();
+
+        /** Sets the pageWidth field. */
+        @JsonProperty("page_width")
+        public Builder withPageWidth(final int value) {
+            this.pageWidth = value;
+            return this;
+        }
+
+        /** Sets the pageHeight field. */
+        @JsonProperty("page_height")
+        public Builder withPageHeight(final int value) {
+            this.pageHeight = value;
+            return this;
+        }
+
+        /** Sets the detections field. */
+        @JsonProperty("detections")
+        public Builder withDetections(final List<LayoutDetection> value) {
+            this.detections = value;
+            return this;
+        }
+
+        /** Builds the DetectionResult instance. */
+        public DetectionResult build() {
+            return new DetectionResult(
+                pageWidth,
+                pageHeight,
+                detections
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/DiffHunk.java

@@ -0,0 +1,106 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * A single contiguous hunk in a unified diff.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = DiffHunk.Builder.class)
+public record DiffHunk(
+    /**
+     * Starting line number in the old content (0-indexed).
+     */
+    @JsonProperty("from_line") long fromLine,
+    /**
+     * Number of lines from the old content in this hunk.
+     */
+    @JsonProperty("from_count") long fromCount,
+    /**
+     * Starting line number in the new content (0-indexed).
+     */
+    @JsonProperty("to_line") long toLine,
+    /**
+     * Number of lines from the new content in this hunk.
+     */
+    @JsonProperty("to_count") long toCount,
+    /**
+     * Lines that make up this hunk.
+     */
+    @JsonProperty("lines") List<DiffLine> lines
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("from_line")
+private long fromLine = 0;
+        @JsonProperty("from_count")
+private long fromCount = 0;
+        @JsonProperty("to_line")
+private long toLine = 0;
+        @JsonProperty("to_count")
+private long toCount = 0;
+private List<DiffLine> lines = List.of();
+
+        /** Sets the fromLine field. */
+        @JsonProperty("from_line")
+        public Builder withFromLine(final long value) {
+            this.fromLine = value;
+            return this;
+        }
+
+        /** Sets the fromCount field. */
+        @JsonProperty("from_count")
+        public Builder withFromCount(final long value) {
+            this.fromCount = value;
+            return this;
+        }
+
+        /** Sets the toLine field. */
+        @JsonProperty("to_line")
+        public Builder withToLine(final long value) {
+            this.toLine = value;
+            return this;
+        }
+
+        /** Sets the toCount field. */
+        @JsonProperty("to_count")
+        public Builder withToCount(final long value) {
+            this.toCount = value;
+            return this;
+        }
+
+        /** Sets the lines field. */
+        @JsonProperty("lines")
+        public Builder withLines(final List<DiffLine> value) {
+            this.lines = value;
+            return this;
+        }
+
+        /** Builds the DiffHunk instance. */
+        public DiffHunk build() {
+            return new DiffHunk(
+                fromLine,
+                fromCount,
+                toLine,
+                toCount,
+                lines
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/DiffLine.java

@@ -0,0 +1,126 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.databind.deser.std.StdDeserializer;
+import com.fasterxml.jackson.databind.ser.std.StdSerializer;
+import com.fasterxml.jackson.core.JsonParser;
+import com.fasterxml.jackson.core.JsonGenerator;
+import com.fasterxml.jackson.databind.DeserializationContext;
+import com.fasterxml.jackson.databind.SerializerProvider;
+import com.fasterxml.jackson.databind.node.ObjectNode;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonSerialize;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * A single line in a unified-diff hunk.
+ *
+ * Defined here (rather than only in {@code crate.diff}) so {@code RevisionDelta} can
+ * reference it unconditionally, without requiring the {@code diff} Cargo feature.
+ * {@code crate.diff} re-exports this type verbatim.
+ */
+@com.fasterxml.jackson.annotation.JsonIgnoreProperties(ignoreUnknown = true)
+@JsonDeserialize(using = DiffLineDeserializer.class)
+@JsonSerialize(using = DiffLineSerializer.class)
+public sealed interface DiffLine {
+
+    /** Unchanged context line. */
+    record Context(String value) implements DiffLine { }
+
+    /** Line added in the "after" version. */
+    record Added(String value) implements DiffLine { }
+
+    /** Line removed from the "before" version. */
+    record Removed(String value) implements DiffLine { }
+
+    /** Returns the Context data if this is a Context variant, otherwise null. */
+    default @Nullable String context() {
+        return this instanceof Context e ? e.value() : null;
+    }
+
+    /** Returns the Added data if this is a Added variant, otherwise null. */
+    default @Nullable String added() {
+        return this instanceof Added e ? e.value() : null;
+    }
+
+    /** Returns the Removed data if this is a Removed variant, otherwise null. */
+    default @Nullable String removed() {
+        return this instanceof Removed e ? e.value() : null;
+    }
+
+}
+
+// Custom deserializer for sealed interface with unwrapped variants
+class DiffLineDeserializer extends StdDeserializer<DiffLine> {
+    DiffLineDeserializer() {
+        super(DiffLine.class);
+    }
+
+    @Override
+    public DiffLine deserialize(JsonParser parser, DeserializationContext ctx)
+            throws java.io.IOException {
+        ObjectNode node = parser.getCodec().readTree(parser);
+        com.fasterxml.jackson.databind.JsonNode tagNode = node.get("kind");
+        if (tagNode == null || tagNode.isNull()) {
+            throw new com.fasterxml.jackson.databind.JsonMappingException(
+                parser, "Missing discriminator field: kind");
+        }
+        String tagValue = tagNode.asText();
+        node.remove("kind");
+
+        return switch (tagValue) {
+            case "context" -> new DiffLine.Context(node.toString());
+            case "added" -> new DiffLine.Added(node.toString());
+            case "removed" -> new DiffLine.Removed(node.toString());
+            default -> throw new com.fasterxml.jackson.databind.JsonMappingException(
+                parser, "Unknown DiffLine discriminator: " + tagValue);
+        };
+    }
+}
+
+// Custom serializer for sealed interface with unwrapped variants — emits
+// the discriminator tag alongside the inner record's fields (flat object).
+class DiffLineSerializer extends StdSerializer<DiffLine> {
+    private static final com.fasterxml.jackson.databind.ObjectMapper MAPPER =
+        new com.fasterxml.jackson.databind.ObjectMapper()
+            .registerModule(new com.fasterxml.jackson.datatype.jdk8.Jdk8Module())
+            .setPropertyNamingStrategy(com.fasterxml.jackson.databind.PropertyNamingStrategies.SNAKE_CASE)
+            .setSerializationInclusion(com.fasterxml.jackson.annotation.JsonInclude.Include.NON_NULL);
+
+    DiffLineSerializer() {
+        super(DiffLine.class);
+    }
+
+    @Override
+    public void serialize(DiffLine value, JsonGenerator gen, SerializerProvider provider)
+            throws java.io.IOException {
+        String tag;
+        Object inner;if (value instanceof DiffLine.Context v) {
+            tag = "context";            inner = v.value();        }else if (value instanceof DiffLine.Added v) {
+            tag = "added";            inner = v.value();        }else if (value instanceof DiffLine.Removed v) {
+            tag = "removed";            inner = v.value();        }        else {
+            throw new com.fasterxml.jackson.databind.JsonMappingException(gen,
+                "Unknown DiffLine variant: " + value.getClass().getName());
+        }
+
+        gen.writeStartObject();
+        gen.writeStringField("kind", tag);
+        if (inner != null) {
+            com.fasterxml.jackson.databind.JsonNode tree = MAPPER.valueToTree(inner);
+            if (tree.isObject()) {
+                java.util.Iterator<java.util.Map.Entry<String, com.fasterxml.jackson.databind.JsonNode>> it =
+                    tree.fields();
+                while (it.hasNext()) {
+                    java.util.Map.Entry<String, com.fasterxml.jackson.databind.JsonNode> e = it.next();
+                    gen.writeFieldName(e.getKey());
+                    gen.writeTree(e.getValue());
+                }
+            }
+        }
+        gen.writeEndObject();
+    }
+}

packages/java/dev/kreuzberg/DiffOptions.java

@@ -0,0 +1,85 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Options controlling how two {@code ExtractionResult} values are compared.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = DiffOptions.Builder.class)
+public record DiffOptions(
+    /**
+     * Include metadata changes in the diff. Default: {@code true}.
+     */
+    @JsonProperty("include_metadata") boolean includeMetadata,
+    /**
+     * Include embedded-children changes in the diff. Default: {@code true}.
+     */
+    @JsonProperty("include_embedded") boolean includeEmbedded,
+    /**
+     * Truncate content to this many characters before diffing.
+     *
+     * Useful for very large documents where only the first N characters matter.
+     * {@code None} means no truncation.
+     */
+    @Nullable @JsonProperty("max_content_chars") Long maxContentChars
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("include_metadata")
+private boolean includeMetadata = true;
+        @JsonProperty("include_embedded")
+private boolean includeEmbedded = true;
+        @JsonProperty("max_content_chars")
+private Long maxContentChars = null;
+
+        /** Sets the includeMetadata field. */
+        @JsonProperty("include_metadata")
+        public Builder withIncludeMetadata(final boolean value) {
+            this.includeMetadata = value;
+            return this;
+        }
+
+        /** Sets the includeEmbedded field. */
+        @JsonProperty("include_embedded")
+        public Builder withIncludeEmbedded(final boolean value) {
+            this.includeEmbedded = value;
+            return this;
+        }
+
+        /** Sets the maxContentChars field. */
+        @JsonProperty("max_content_chars")
+        public Builder withMaxContentChars(final @Nullable long value) {
+            this.maxContentChars = value;
+            return this;
+        }
+
+        /** Builds the DiffOptions instance. */
+        public DiffOptions build() {
+            return new DiffOptions(
+                includeMetadata,
+                includeEmbedded,
+                maxContentChars
+            );
+        }
+    }
+    // CPD-ON
+    public static DiffOptions defaultInstance() {
+        throw new UnsupportedOperationException("defaultInstance is not yet bridged via JNI; use the Builder instead.");
+    }
+}

packages/java/dev/kreuzberg/DjotContent.java

@@ -0,0 +1,153 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Comprehensive Djot document structure with semantic preservation.
+ *
+ * This type captures the full richness of Djot markup, including:
+ * - Block-level structures (headings, lists, blockquotes, code blocks, etc.)
+ * - Inline formatting (emphasis, strong, highlight, subscript, superscript, etc.)
+ * - Attributes (classes, IDs, key-value pairs)
+ * - Links, images, footnotes
+ * - Math expressions (inline and display)
+ * - Tables with full structure
+ *
+ * Available when the {@code djot} feature is enabled.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = DjotContent.Builder.class)
+public record DjotContent(
+    /**
+     * Plain text representation for backwards compatibility
+     */
+    @JsonProperty("plain_text") String plainText,
+    /**
+     * Structured block-level content
+     */
+    @JsonProperty("blocks") List<FormattedBlock> blocks,
+    /**
+     * Metadata from YAML frontmatter
+     */
+    @JsonProperty("metadata") Metadata metadata,
+    /**
+     * Extracted tables as structured data
+     */
+    @JsonProperty("tables") List<Table> tables,
+    /**
+     * Extracted images with metadata
+     */
+    @JsonProperty("images") List<DjotImage> images,
+    /**
+     * Extracted links with URLs
+     */
+    @JsonProperty("links") List<DjotLink> links,
+    /**
+     * Footnote definitions
+     */
+    @JsonProperty("footnotes") List<Footnote> footnotes,
+    /**
+     * Attributes mapped by element identifier (if present)
+     */
+    @Nullable @JsonProperty("attributes") List<String> attributes
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("plain_text")
+private String plainText = "";
+private List<FormattedBlock> blocks = List.of();
+private Metadata metadata = null;
+private List<Table> tables = List.of();
+private List<DjotImage> images = List.of();
+private List<DjotLink> links = List.of();
+private List<Footnote> footnotes = List.of();
+private List<String> attributes = null;
+
+        /** Sets the plainText field. */
+        @JsonProperty("plain_text")
+        public Builder withPlainText(final String value) {
+            this.plainText = value;
+            return this;
+        }
+
+        /** Sets the blocks field. */
+        @JsonProperty("blocks")
+        public Builder withBlocks(final List<FormattedBlock> value) {
+            this.blocks = value;
+            return this;
+        }
+
+        /** Sets the metadata field. */
+        @JsonProperty("metadata")
+        public Builder withMetadata(final Metadata value) {
+            this.metadata = value;
+            return this;
+        }
+
+        /** Sets the tables field. */
+        @JsonProperty("tables")
+        public Builder withTables(final List<Table> value) {
+            this.tables = value;
+            return this;
+        }
+
+        /** Sets the images field. */
+        @JsonProperty("images")
+        public Builder withImages(final List<DjotImage> value) {
+            this.images = value;
+            return this;
+        }
+
+        /** Sets the links field. */
+        @JsonProperty("links")
+        public Builder withLinks(final List<DjotLink> value) {
+            this.links = value;
+            return this;
+        }
+
+        /** Sets the footnotes field. */
+        @JsonProperty("footnotes")
+        public Builder withFootnotes(final List<Footnote> value) {
+            this.footnotes = value;
+            return this;
+        }
+
+        /** Sets the attributes field. */
+        @JsonProperty("attributes")
+        public Builder withAttributes(final @Nullable List<String> value) {
+            this.attributes = value;
+            return this;
+        }
+
+        /** Builds the DjotContent instance. */
+        public DjotContent build() {
+            return new DjotContent(
+                plainText,
+                blocks,
+                metadata,
+                tables,
+                images,
+                links,
+                footnotes,
+                attributes
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/DjotImage.java

@@ -0,0 +1,89 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Image element in Djot.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = DjotImage.Builder.class)
+public record DjotImage(
+    /**
+     * Image source URL or path
+     */
+    @JsonProperty("src") String src,
+    /**
+     * Alternative text
+     */
+    @JsonProperty("alt") String alt,
+    /**
+     * Optional title
+     */
+    @Nullable @JsonProperty("title") String title,
+    /**
+     * Element attributes
+     */
+    @Nullable @JsonProperty("attributes") String attributes
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String src = "";
+private String alt = "";
+private String title = null;
+private String attributes = null;
+
+        /** Sets the src field. */
+        @JsonProperty("src")
+        public Builder withSrc(final String value) {
+            this.src = value;
+            return this;
+        }
+
+        /** Sets the alt field. */
+        @JsonProperty("alt")
+        public Builder withAlt(final String value) {
+            this.alt = value;
+            return this;
+        }
+
+        /** Sets the title field. */
+        @JsonProperty("title")
+        public Builder withTitle(final @Nullable String value) {
+            this.title = value;
+            return this;
+        }
+
+        /** Sets the attributes field. */
+        @JsonProperty("attributes")
+        public Builder withAttributes(final @Nullable String value) {
+            this.attributes = value;
+            return this;
+        }
+
+        /** Builds the DjotImage instance. */
+        public DjotImage build() {
+            return new DjotImage(
+                src,
+                alt,
+                title,
+                attributes
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/DjotLink.java

@@ -0,0 +1,89 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Link element in Djot.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = DjotLink.Builder.class)
+public record DjotLink(
+    /**
+     * Link URL
+     */
+    @JsonProperty("url") String url,
+    /**
+     * Link text content
+     */
+    @JsonProperty("text") String text,
+    /**
+     * Optional title
+     */
+    @Nullable @JsonProperty("title") String title,
+    /**
+     * Element attributes
+     */
+    @Nullable @JsonProperty("attributes") String attributes
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String url = "";
+private String text = "";
+private String title = null;
+private String attributes = null;
+
+        /** Sets the url field. */
+        @JsonProperty("url")
+        public Builder withUrl(final String value) {
+            this.url = value;
+            return this;
+        }
+
+        /** Sets the text field. */
+        @JsonProperty("text")
+        public Builder withText(final String value) {
+            this.text = value;
+            return this;
+        }
+
+        /** Sets the title field. */
+        @JsonProperty("title")
+        public Builder withTitle(final @Nullable String value) {
+            this.title = value;
+            return this;
+        }
+
+        /** Sets the attributes field. */
+        @JsonProperty("attributes")
+        public Builder withAttributes(final @Nullable String value) {
+            this.attributes = value;
+            return this;
+        }
+
+        /** Builds the DjotLink instance. */
+        public DjotLink build() {
+            return new DjotLink(
+                url,
+                text,
+                title,
+                attributes
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/DocumentExtractorBridge.java

@@ -0,0 +1,402 @@
+package dev.kreuzberg;
+
+import java.lang.foreign.Arena;
+import java.lang.foreign.FunctionDescriptor;
+import java.lang.foreign.Linker;
+import java.lang.foreign.MemoryLayout;
+import java.lang.foreign.MemorySegment;
+import java.lang.foreign.ValueLayout;
+import java.lang.invoke.MethodHandles;
+import java.lang.invoke.MethodType;
+import java.util.List;
+import java.util.concurrent.ConcurrentHashMap;
+import com.fasterxml.jackson.databind.ObjectMapper;
+
+/**
+ * Allocates Panama FFM upcall stubs for an IDocumentExtractor implementation,
+ * assembles the C vtable in native memory, and provides static
+ * registerDocumentExtractor/unregisterDocumentExtractor helpers.
+ */
+public final class DocumentExtractorBridge implements AutoCloseable {
+
+    private static final Linker LINKER = Linker.nativeLinker();
+    private static final MethodHandles.Lookup LOOKUP = MethodHandles.lookup();
+    private static final ObjectMapper JSON = new ObjectMapper();
+
+    /** Live registry — keeps Arenas and upcall stubs alive past the register call. */
+    private static final ConcurrentHashMap<String, DocumentExtractorBridge>
+            DOCUMENT_EXTRACTOR_BRIDGES = new ConcurrentHashMap<>();
+
+    // C vtable: 11 fields (4 plugin methods + 5 trait methods + free_string + free_user_data)
+    private static final MemoryLayout VTABLE_LAYOUT = MemoryLayout.structLayout(ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS);
+    private static final long VTABLE_SIZE = VTABLE_LAYOUT.byteSize();
+
+    private final Arena arena;
+    private final MemorySegment vtable;
+    private final IDocumentExtractor impl;
+
+    DocumentExtractorBridge(final IDocumentExtractor impl) {
+        this.impl = impl;
+        this.arena = Arena.ofShared();
+        this.vtable = arena.allocate(VTABLE_SIZE);
+
+        try {
+            long offset = 0L;
+
+            var stubName = LINKER.upcallStub(LOOKUP.bind(this, "handleName",
+                MethodType.methodType(int.class, MemorySegment.class, MemorySegment.class, MemorySegment.class)),
+                FunctionDescriptor.of(ValueLayout.JAVA_INT, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubName);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+            var stubVersion = LINKER.upcallStub(LOOKUP.bind(this, "handleVersion",
+                MethodType.methodType(int.class, MemorySegment.class, MemorySegment.class, MemorySegment.class)),
+                FunctionDescriptor.of(ValueLayout.JAVA_INT, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubVersion);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+            var stubInitialize = LINKER.upcallStub(LOOKUP.bind(this, "handleInitialize",
+                MethodType.methodType(int.class, MemorySegment.class, MemorySegment.class)),
+                FunctionDescriptor.of(ValueLayout.JAVA_INT, ValueLayout.ADDRESS, ValueLayout.ADDRESS),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubInitialize);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+            var stubShutdown = LINKER.upcallStub(LOOKUP.bind(this, "handleShutdown",
+                MethodType.methodType(int.class, MemorySegment.class, MemorySegment.class)),
+                FunctionDescriptor.of(ValueLayout.JAVA_INT, ValueLayout.ADDRESS, ValueLayout.ADDRESS),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubShutdown);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+            var stubExtractBytes = LINKER.upcallStub(LOOKUP.bind(this, "handleExtractBytes",
+                MethodType.methodType(
+                    int.class,
+                    MemorySegment.class,
+                    MemorySegment.class,
+                    long.class,
+                    MemorySegment.class,
+                    MemorySegment.class,
+                    MemorySegment.class,
+                    MemorySegment.class
+                )),
+                FunctionDescriptor.of(
+                    ValueLayout.JAVA_INT,
+                    ValueLayout.ADDRESS,
+                    ValueLayout.ADDRESS,
+                    ValueLayout.JAVA_LONG,
+                    ValueLayout.ADDRESS,
+                    ValueLayout.ADDRESS,
+                    ValueLayout.ADDRESS,
+                    ValueLayout.ADDRESS
+                ),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubExtractBytes);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+            var stubExtractFile = LINKER.upcallStub(LOOKUP.bind(this, "handleExtractFile",
+                MethodType.methodType(
+                    int.class,
+                    MemorySegment.class,
+                    MemorySegment.class,
+                    MemorySegment.class,
+                    MemorySegment.class,
+                    MemorySegment.class,
+                    MemorySegment.class
+                )),
+                FunctionDescriptor.of(
+                    ValueLayout.JAVA_INT,
+                    ValueLayout.ADDRESS,
+                    ValueLayout.ADDRESS,
+                    ValueLayout.ADDRESS,
+                    ValueLayout.ADDRESS,
+                    ValueLayout.ADDRESS,
+                    ValueLayout.ADDRESS
+                ),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubExtractFile);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+            var stubSupportedMimeTypes = LINKER.upcallStub(LOOKUP.bind(this, "handleSupportedMimeTypes",
+                MethodType.methodType(int.class, MemorySegment.class, MemorySegment.class, MemorySegment.class)),
+                FunctionDescriptor.of(ValueLayout.JAVA_INT, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubSupportedMimeTypes);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+            var stubPriority = LINKER.upcallStub(LOOKUP.bind(this, "handlePriority",
+                MethodType.methodType(int.class, MemorySegment.class, MemorySegment.class, MemorySegment.class)),
+                FunctionDescriptor.of(ValueLayout.JAVA_INT, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubPriority);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+            var stubCanHandle = LINKER.upcallStub(LOOKUP.bind(this, "handleCanHandle",
+                MethodType.methodType(
+                    int.class,
+                    MemorySegment.class,
+                    MemorySegment.class,
+                    MemorySegment.class,
+                    MemorySegment.class,
+                    MemorySegment.class
+                )),
+                FunctionDescriptor.of(
+                    ValueLayout.JAVA_INT,
+                    ValueLayout.ADDRESS,
+                    ValueLayout.ADDRESS,
+                    ValueLayout.ADDRESS,
+                    ValueLayout.ADDRESS,
+                    ValueLayout.ADDRESS
+                ),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubCanHandle);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+            var stubFreeString = LINKER.upcallStub(LOOKUP.bind(this, "freeString",
+                MethodType.methodType(void.class, MemorySegment.class)),
+                FunctionDescriptor.ofVoid(ValueLayout.ADDRESS),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubFreeString);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+            var stubFreeUserData = LINKER.upcallStub(LOOKUP.bind(this, "freeUserData",
+                MethodType.methodType(void.class, MemorySegment.class)),
+                FunctionDescriptor.ofVoid(ValueLayout.ADDRESS),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubFreeUserData);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+
+        } catch (ReflectiveOperationException e) {
+            arena.close();
+            throw new RuntimeException("Failed to create trait bridge stubs", e);
+        }
+    }
+
+    MemorySegment vtableSegment() { return vtable; }
+
+    private int handleName(MemorySegment userData, MemorySegment outName, MemorySegment outError) {
+        try {
+            outName.set(ValueLayout.ADDRESS, 0, arena.allocateFrom(impl.name()));
+            return 0;
+        } catch (Throwable e) { return 1; }
+    }
+
+    private int handleVersion(MemorySegment userData, MemorySegment outVersion, MemorySegment outError) {
+        try {
+            outVersion.set(ValueLayout.ADDRESS, 0, arena.allocateFrom(impl.version()));
+            return 0;
+        } catch (Throwable e) { return 1; }
+    }
+
+    private int handleInitialize(MemorySegment userData, MemorySegment outError) {
+        try {
+            impl.initialize();
+            return 0;
+        } catch (Throwable e) { return 1; }
+    }
+
+    private int handleShutdown(MemorySegment userData, MemorySegment outError) {
+        try {
+            impl.shutdown();
+            return 0;
+        } catch (Throwable e) { return 1; }
+    }
+
+    private int handleExtractBytes(
+        MemorySegment userData,
+        MemorySegment content_in,
+        long contentLen,
+        MemorySegment mime_type_in,
+        MemorySegment config_in,
+        MemorySegment outResult,
+        MemorySegment outError
+    ) {
+        try {
+            byte[] content = content_in.reinterpret(contentLen).toArray(ValueLayout.JAVA_BYTE);
+            String mime_type = mime_type_in.reinterpret(Long.MAX_VALUE).getString(0);
+            String config_json = config_in.reinterpret(Long.MAX_VALUE).getString(0);
+            ExtractionConfig config = JSON.readValue(config_json, ExtractionConfig.class);
+            String result = impl.extract_bytes(content, mime_type, config);
+            MemorySegment jsonCs = arena.allocateFrom(result);
+            outResult.set(ValueLayout.ADDRESS, 0, jsonCs);
+            return 0;
+        } catch (Throwable e) {
+            writeError(outError, e);
+            return 1;
+        }
+    }
+
+    private int handleExtractFile(
+        MemorySegment userData,
+        MemorySegment path_in,
+        MemorySegment mime_type_in,
+        MemorySegment config_in,
+        MemorySegment outResult,
+        MemorySegment outError
+    ) {
+        try {
+            java.nio.file.Path path = java.nio.file.Paths.get(path_in.reinterpret(Long.MAX_VALUE).getString(0));
+            String mime_type = mime_type_in.reinterpret(Long.MAX_VALUE).getString(0);
+            String config_json = config_in.reinterpret(Long.MAX_VALUE).getString(0);
+            ExtractionConfig config = JSON.readValue(config_json, ExtractionConfig.class);
+            String result = impl.extract_file(path, mime_type, config);
+            MemorySegment jsonCs = arena.allocateFrom(result);
+            outResult.set(ValueLayout.ADDRESS, 0, jsonCs);
+            return 0;
+        } catch (Throwable e) {
+            writeError(outError, e);
+            return 1;
+        }
+    }
+
+    private int handleSupportedMimeTypes(MemorySegment userData, MemorySegment outResult, MemorySegment outError) {
+        try {
+            List<String> result = impl.supported_mime_types();
+            String json = JSON.writeValueAsString(result);
+            MemorySegment jsonCs = arena.allocateFrom(json);
+            outResult.set(ValueLayout.ADDRESS, 0, jsonCs);
+            return 0;
+        } catch (Throwable e) {
+            writeError(outError, e);
+            return 1;
+        }
+    }
+
+    private int handlePriority(MemorySegment userData, MemorySegment outResult, MemorySegment outError) {
+        try {
+            int result = impl.priority();
+            String json = JSON.writeValueAsString(result);
+            MemorySegment jsonCs = arena.allocateFrom(json);
+            outResult.set(ValueLayout.ADDRESS, 0, jsonCs);
+            return 0;
+        } catch (Throwable e) {
+            writeError(outError, e);
+            return 1;
+        }
+    }
+
+    private int handleCanHandle(
+        MemorySegment userData,
+        MemorySegment _path_in,
+        MemorySegment _mime_type_in,
+        MemorySegment outResult,
+        MemorySegment outError
+    ) {
+        try {
+            java.nio.file.Path _path = java.nio.file.Paths.get(_path_in.reinterpret(Long.MAX_VALUE).getString(0));
+            String _mime_type = _mime_type_in.reinterpret(Long.MAX_VALUE).getString(0);
+            boolean result = impl.can_handle(_path, _mime_type);
+            String json = JSON.writeValueAsString(result);
+            MemorySegment jsonCs = arena.allocateFrom(json);
+            outResult.set(ValueLayout.ADDRESS, 0, jsonCs);
+            return 0;
+        } catch (Throwable e) {
+            writeError(outError, e);
+            return 1;
+        }
+    }
+
+    private void writeError(MemorySegment outError, Throwable e) {
+        try { outError.set(ValueLayout.ADDRESS, 0, arena.allocateFrom(e.getClass().getSimpleName() + ": " + e.getMessage())); }
+        catch (Throwable ignored) { /* swallow */ }
+    }
+
+    private void freeString(MemorySegment ptr) {
+        // Strings returned by Java callbacks are arena-owned and released when this bridge closes.
+    }
+
+    private void freeUserData(MemorySegment userData) {
+        // User data is Java-side state (the impl object), not freed by Rust on drop.
+    }
+
+    /** Read a NUL-terminated native C string safely without unbounded reinterpret. */
+    private static String readNativeString(MemorySegment ptr) {
+        return ptr.reinterpret(4096).getString(0);
+    }
+
+    @Override
+    public void close() { arena.close(); }
+
+    /** Register a DocumentExtractor implementation via Panama FFM upcall stubs. */
+    public static void registerDocumentExtractor(final IDocumentExtractor impl) throws Exception {
+        var bridge = new DocumentExtractorBridge(impl);
+        try {
+            try (var nameArena = Arena.ofShared()) {
+                var nameCs = nameArena.allocateFrom(impl.name());
+                MemorySegment outErr = nameArena.allocate(ValueLayout.ADDRESS);
+                int rc = (int) NativeLib.KREUZBERG_REGISTER_DOCUMENT_EXTRACTOR.invoke(
+                    nameCs,
+                    bridge.vtableSegment(),
+                    MemorySegment.NULL,
+                    outErr
+                );
+                if (rc != 0) {
+                    MemorySegment errPtr = outErr.get(ValueLayout.ADDRESS, 0);
+                    String msg = errPtr.equals(MemorySegment.NULL) ? "registration failed (rc=" + rc + ")" : readNativeString(errPtr);
+                    throw new RuntimeException("registerDocumentExtractor: " + msg);
+                }
+            }
+        } catch (Throwable t) {
+            bridge.close();
+            if (t instanceof Exception e) {
+                throw e;
+            } else {
+                throw new RuntimeException("Unexpected error during registration", t);
+            }
+        }
+        DOCUMENT_EXTRACTOR_BRIDGES.put(impl.name(), bridge);
+    }
+
+    /** Unregister a DocumentExtractor implementation by name. */
+    public static void unregisterDocumentExtractor(String name) throws Exception {
+        try {
+            try (var nameArena = Arena.ofShared()) {
+                var nameCs = nameArena.allocateFrom(name);
+                MemorySegment outErr = nameArena.allocate(ValueLayout.ADDRESS);
+                int rc = (int) NativeLib.KREUZBERG_UNREGISTER_DOCUMENT_EXTRACTOR.invoke(nameCs, outErr);
+                if (rc != 0) {
+                    MemorySegment errPtr = outErr.get(ValueLayout.ADDRESS, 0);
+                    String msg = errPtr.equals(MemorySegment.NULL)
+                        ? "unregistration failed (rc=" + rc + ")"
+                        : errPtr.reinterpret(Long.MAX_VALUE).getString(0);
+                    throw new RuntimeException("unregisterDocumentExtractor: " + msg);
+                }
+            }
+        } catch (Throwable t) {
+            if (t instanceof Exception e) {
+                throw e;
+            } else {
+                throw new RuntimeException("Unexpected error during unregistration", t);
+            }
+        }
+        DocumentExtractorBridge old = DOCUMENT_EXTRACTOR_BRIDGES.remove(name);
+        if (old != null) { old.close(); }
+    }
+    /** Clear all registered DocumentExtractor implementations. */
+    public static void clearDocumentExtractors() throws Exception {
+        try {
+            try (var arena = Arena.ofShared()) {
+                MemorySegment outErr = arena.allocate(ValueLayout.ADDRESS);
+                int rc = (int) NativeLib.KREUZBERG_CLEAR_DOCUMENT_EXTRACTOR.invoke(outErr);
+                if (rc != 0) {
+                    MemorySegment errPtr = outErr.get(ValueLayout.ADDRESS, 0);
+                    String msg = errPtr.equals(MemorySegment.NULL)
+                        ? "clear failed (rc=" + rc + ")"
+                        : errPtr.reinterpret(Long.MAX_VALUE).getString(0);
+                    throw new RuntimeException("clearDocumentExtractors: " + msg);
+                }
+            }
+        } catch (Throwable t) {
+            if (t instanceof Exception e) {
+                throw e;
+            } else {
+                throw new RuntimeException("Unexpected error during clear", t);
+            }
+        }
+        DOCUMENT_EXTRACTOR_BRIDGES.values().forEach(DocumentExtractorBridge::close);
+        DOCUMENT_EXTRACTOR_BRIDGES.clear();
+    }
+}

packages/java/dev/kreuzberg/DocumentNode.java

@@ -0,0 +1,179 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import java.util.Map;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * A single node in the document tree.
+ *
+ * Each node has deterministic {@code id}, typed {@code content}, optional {@code parent}/{@code children}
+ * for tree structure, and metadata like page number, bounding box, and content layer.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = DocumentNode.Builder.class)
+public record DocumentNode(
+    /**
+     * Deterministic identifier (hash of content + position).
+     */
+    @JsonProperty("id") String id,
+    /**
+     * Node content — tagged enum, type-specific data only.
+     */
+    @JsonProperty("content") NodeContent content,
+    /**
+     * Parent node index ({@code None} = root-level node).
+     */
+    @Nullable @JsonProperty("parent") Integer parent,
+    /**
+     * Child node indices in reading order.
+     */
+    @Nullable @JsonProperty("children") List<Integer> children,
+    /**
+     * Content layer classification.
+     */
+    @Nullable @JsonProperty("content_layer") ContentLayer contentLayer,
+    /**
+     * Page number where this node starts (1-indexed).
+     */
+    @Nullable @JsonProperty("page") Integer page,
+    /**
+     * Page number where this node ends (for multi-page tables/sections).
+     */
+    @Nullable @JsonProperty("page_end") Integer pageEnd,
+    /**
+     * Bounding box in document coordinates.
+     */
+    @Nullable @JsonProperty("bbox") BoundingBox bbox,
+    /**
+     * Inline annotations (formatting, links) on this node's text content.
+     *
+     * Only meaningful for text-carrying nodes; empty for containers.
+     */
+    @Nullable @JsonProperty("annotations") List<TextAnnotation> annotations,
+    /**
+     * Format-specific key-value attributes.
+     *
+     * Extensible bag for miscellaneous data without a dedicated typed field: CSS classes,
+     * LaTeX environment names, Excel cell formulas, slide layout names, etc.
+     */
+    @Nullable @JsonProperty("attributes") Map<String, String> attributes
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String id = "";
+private NodeContent content = null;
+private Integer parent = null;
+private List<Integer> children = null;
+        @JsonProperty("content_layer")
+        @Nullable private ContentLayer contentLayer = ContentLayer.Body;
+private Integer page = null;
+        @JsonProperty("page_end")
+private Integer pageEnd = null;
+private BoundingBox bbox = null;
+private List<TextAnnotation> annotations = null;
+private Map<String, String> attributes = null;
+
+        /** Sets the id field. */
+        @JsonProperty("id")
+        public Builder withId(final String value) {
+            this.id = value;
+            return this;
+        }
+
+        /** Sets the content field. */
+        @JsonProperty("content")
+        public Builder withContent(final NodeContent value) {
+            this.content = value;
+            return this;
+        }
+
+        /** Sets the parent field. */
+        @JsonProperty("parent")
+        public Builder withParent(final @Nullable int value) {
+            this.parent = value;
+            return this;
+        }
+
+        /** Sets the children field. */
+        @JsonProperty("children")
+        public Builder withChildren(final @Nullable List<Integer> value) {
+            this.children = value;
+            return this;
+        }
+
+        /** Sets the contentLayer field. */
+        @JsonProperty("content_layer")
+        public Builder withContentLayer(final @Nullable ContentLayer value) {
+            this.contentLayer = value;
+            return this;
+        }
+
+        /** Sets the page field. */
+        @JsonProperty("page")
+        public Builder withPage(final @Nullable int value) {
+            this.page = value;
+            return this;
+        }
+
+        /** Sets the pageEnd field. */
+        @JsonProperty("page_end")
+        public Builder withPageEnd(final @Nullable int value) {
+            this.pageEnd = value;
+            return this;
+        }
+
+        /** Sets the bbox field. */
+        @JsonProperty("bbox")
+        public Builder withBbox(final @Nullable BoundingBox value) {
+            this.bbox = value;
+            return this;
+        }
+
+        /** Sets the annotations field. */
+        @JsonProperty("annotations")
+        public Builder withAnnotations(final @Nullable List<TextAnnotation> value) {
+            this.annotations = value;
+            return this;
+        }
+
+        /** Sets the attributes field. */
+        @JsonProperty("attributes")
+        public Builder withAttributes(final @Nullable Map<String, String> value) {
+            this.attributes = value;
+            return this;
+        }
+
+        /** Builds the DocumentNode instance. */
+        public DocumentNode build() {
+            return new DocumentNode(
+                id,
+                content,
+                parent,
+                children,
+                contentLayer,
+                page,
+                pageEnd,
+                bbox,
+                annotations,
+                attributes
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/DocumentRelationship.java

@@ -0,0 +1,75 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * A resolved relationship between two nodes in the document tree.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = DocumentRelationship.Builder.class)
+public record DocumentRelationship(
+    /**
+     * Source node index (the referencing node).
+     */
+    @JsonProperty("source") int source,
+    /**
+     * Target node index (the referenced node).
+     */
+    @JsonProperty("target") int target,
+    /**
+     * Semantic kind of the relationship.
+     */
+    @JsonProperty("kind") RelationshipKind kind
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private int source = 0;
+private int target = 0;
+private RelationshipKind kind = null;
+
+        /** Sets the source field. */
+        @JsonProperty("source")
+        public Builder withSource(final int value) {
+            this.source = value;
+            return this;
+        }
+
+        /** Sets the target field. */
+        @JsonProperty("target")
+        public Builder withTarget(final int value) {
+            this.target = value;
+            return this;
+        }
+
+        /** Sets the kind field. */
+        @JsonProperty("kind")
+        public Builder withKind(final RelationshipKind value) {
+            this.kind = value;
+            return this;
+        }
+
+        /** Builds the DocumentRelationship instance. */
+        public DocumentRelationship build() {
+            return new DocumentRelationship(
+                source,
+                target,
+                kind
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/DocumentRevision.java

@@ -0,0 +1,134 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * A single tracked change embedded in a document.
+ *
+ * Populated by per-format extractors that understand change-tracking metadata
+ * (DOCX {@code w:ins}/{@code w:del}/{@code w:rPrChange}, ODT {@code text:change-*}, …). Every
+ * extractor defaults to {@code ExtractionResult.revisions = None} until a
+ * format-specific implementation is added.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = DocumentRevision.Builder.class)
+public record DocumentRevision(
+    /**
+     * Format-specific revision identifier.
+     *
+     * For DOCX this is the {@code w:id} attribute value on the change element
+     * (e.g. {@code "42"}). When the attribute is absent a synthetic fallback is
+     * generated ({@code "docx-ins-0"}, {@code "docx-del-3"}, …).
+     */
+    @JsonProperty("revision_id") String revisionId,
+    /**
+     * Display name of the author who made this change, when available.
+     */
+    @Nullable @JsonProperty("author") String author,
+    /**
+     * ISO-8601 timestamp of the change, when available.
+     *
+     * Stored as a plain string so this type remains FFI-friendly and
+     * unconditionally available without the {@code chrono} optional dep.
+     * DOCX populates this from the {@code w:date} attribute (e.g.
+     * {@code "2024-03-15T10:30:00Z"}).
+     */
+    @Nullable @JsonProperty("timestamp") String timestamp,
+    /**
+     * Semantic kind of this revision.
+     */
+    @JsonProperty("kind") RevisionKind kind,
+    /**
+     * Best-effort document location for this revision.
+     *
+     * Resolution is format-dependent and may be {@code None} when the location
+     * cannot be determined (e.g. changes inside table cells before
+     * table-cell anchor support is added).
+     */
+    @Nullable @JsonProperty("anchor") RevisionAnchor anchor,
+    /**
+     * The content changes that make up this revision.
+     */
+    @JsonProperty("delta") RevisionDelta delta
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("revision_id")
+private String revisionId = "";
+private String author = null;
+private String timestamp = null;
+private RevisionKind kind = null;
+private RevisionAnchor anchor = null;
+private RevisionDelta delta = null;
+
+        /** Sets the revisionId field. */
+        @JsonProperty("revision_id")
+        public Builder withRevisionId(final String value) {
+            this.revisionId = value;
+            return this;
+        }
+
+        /** Sets the author field. */
+        @JsonProperty("author")
+        public Builder withAuthor(final @Nullable String value) {
+            this.author = value;
+            return this;
+        }
+
+        /** Sets the timestamp field. */
+        @JsonProperty("timestamp")
+        public Builder withTimestamp(final @Nullable String value) {
+            this.timestamp = value;
+            return this;
+        }
+
+        /** Sets the kind field. */
+        @JsonProperty("kind")
+        public Builder withKind(final RevisionKind value) {
+            this.kind = value;
+            return this;
+        }
+
+        /** Sets the anchor field. */
+        @JsonProperty("anchor")
+        public Builder withAnchor(final @Nullable RevisionAnchor value) {
+            this.anchor = value;
+            return this;
+        }
+
+        /** Sets the delta field. */
+        @JsonProperty("delta")
+        public Builder withDelta(final RevisionDelta value) {
+            this.delta = value;
+            return this;
+        }
+
+        /** Builds the DocumentRevision instance. */
+        public DocumentRevision build() {
+            return new DocumentRevision(
+                revisionId,
+                author,
+                timestamp,
+                kind,
+                anchor,
+                delta
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/DocumentStructure.java

@@ -0,0 +1,117 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Top-level structured document representation.
+ *
+ * A flat array of nodes with index-based parent/child references forming a tree.
+ * Root-level nodes have {@code parent: None}. Use {@code body_roots()} and {@code furniture_roots()}
+ * to iterate over top-level content by layer.
+ *
+ * # Validation
+ *
+ * Call {@code validate()} after construction to verify all node indices are in bounds
+ * and parent-child relationships are bidirectionally consistent.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = DocumentStructure.Builder.class)
+public record DocumentStructure(
+    /**
+     * All nodes in document/reading order.
+     */
+    @JsonProperty("nodes") List<DocumentNode> nodes,
+    /**
+     * Origin format identifier (e.g. "docx", "pptx", "html", "pdf").
+     *
+     * Allows renderers to apply format-aware heuristics when converting
+     * the document tree to output formats.
+     */
+    @Nullable @JsonProperty("source_format") String sourceFormat,
+    /**
+     * Resolved relationships between nodes (footnote refs, citations, anchor links, etc.).
+     *
+     * Populated during derivation from the internal document representation.
+     * Empty when no relationships are detected.
+     */
+    @Nullable @JsonProperty("relationships") List<DocumentRelationship> relationships,
+    /**
+     * Sorted, deduplicated list of node type names present in this document.
+     *
+     * Each value is the snake_case {@code node_type} tag of the corresponding
+     * NodeContent variant (e.g. {@code "paragraph"}, {@code "heading"}, {@code "table"}, …).
+     *
+     * Computed from nodes via DocumentStructure.finalize_node_types.
+     * Empty until that method is called (internal construction paths call it
+     * at the end of derivation).
+     */
+    @Nullable @JsonProperty("node_types") List<String> nodeTypes
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private List<DocumentNode> nodes = List.of();
+        @JsonProperty("source_format")
+private String sourceFormat = null;
+private List<DocumentRelationship> relationships = null;
+        @JsonProperty("node_types")
+private List<String> nodeTypes = null;
+
+        /** Sets the nodes field. */
+        @JsonProperty("nodes")
+        public Builder withNodes(final List<DocumentNode> value) {
+            this.nodes = value;
+            return this;
+        }
+
+        /** Sets the sourceFormat field. */
+        @JsonProperty("source_format")
+        public Builder withSourceFormat(final @Nullable String value) {
+            this.sourceFormat = value;
+            return this;
+        }
+
+        /** Sets the relationships field. */
+        @JsonProperty("relationships")
+        public Builder withRelationships(final @Nullable List<DocumentRelationship> value) {
+            this.relationships = value;
+            return this;
+        }
+
+        /** Sets the nodeTypes field. */
+        @JsonProperty("node_types")
+        public Builder withNodeTypes(final @Nullable List<String> value) {
+            this.nodeTypes = value;
+            return this;
+        }
+
+        /** Builds the DocumentStructure instance. */
+        public DocumentStructure build() {
+            return new DocumentStructure(
+                nodes,
+                sourceFormat,
+                relationships,
+                nodeTypes
+            );
+        }
+    }
+    // CPD-ON
+    public static DocumentStructure defaultInstance() {
+        throw new UnsupportedOperationException("defaultInstance is not yet bridged via JNI; use the Builder instead.");
+    }
+}

packages/java/dev/kreuzberg/DocxAppProperties.java

@@ -0,0 +1,255 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Application properties from docProps/app.xml for DOCX
+ *
+ * Contains Word-specific document statistics and metadata.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = DocxAppProperties.Builder.class)
+public record DocxAppProperties(
+    /**
+     * Application name (e.g., "Microsoft Office Word")
+     */
+    @Nullable @JsonProperty("application") String application,
+    /**
+     * Application version
+     */
+    @Nullable @JsonProperty("app_version") String appVersion,
+    /**
+     * Template filename
+     */
+    @Nullable @JsonProperty("template") String template,
+    /**
+     * Total editing time in minutes
+     */
+    @Nullable @JsonProperty("total_time") Integer totalTime,
+    /**
+     * Number of pages
+     */
+    @Nullable @JsonProperty("pages") Integer pages,
+    /**
+     * Number of words
+     */
+    @Nullable @JsonProperty("words") Integer words,
+    /**
+     * Number of characters (excluding spaces)
+     */
+    @Nullable @JsonProperty("characters") Integer characters,
+    /**
+     * Number of characters (including spaces)
+     */
+    @Nullable @JsonProperty("characters_with_spaces") Integer charactersWithSpaces,
+    /**
+     * Number of lines
+     */
+    @Nullable @JsonProperty("lines") Integer lines,
+    /**
+     * Number of paragraphs
+     */
+    @Nullable @JsonProperty("paragraphs") Integer paragraphs,
+    /**
+     * Company name
+     */
+    @Nullable @JsonProperty("company") String company,
+    /**
+     * Document security level
+     */
+    @Nullable @JsonProperty("doc_security") Integer docSecurity,
+    /**
+     * Scale crop flag
+     */
+    @Nullable @JsonProperty("scale_crop") Boolean scaleCrop,
+    /**
+     * Links up to date flag
+     */
+    @Nullable @JsonProperty("links_up_to_date") Boolean linksUpToDate,
+    /**
+     * Shared document flag
+     */
+    @Nullable @JsonProperty("shared_doc") Boolean sharedDoc,
+    /**
+     * Hyperlinks changed flag
+     */
+    @Nullable @JsonProperty("hyperlinks_changed") Boolean hyperlinksChanged
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String application = null;
+        @JsonProperty("app_version")
+private String appVersion = null;
+private String template = null;
+        @JsonProperty("total_time")
+private Integer totalTime = null;
+private Integer pages = null;
+private Integer words = null;
+private Integer characters = null;
+        @JsonProperty("characters_with_spaces")
+private Integer charactersWithSpaces = null;
+private Integer lines = null;
+private Integer paragraphs = null;
+private String company = null;
+        @JsonProperty("doc_security")
+private Integer docSecurity = null;
+        @JsonProperty("scale_crop")
+private Boolean scaleCrop = null;
+        @JsonProperty("links_up_to_date")
+private Boolean linksUpToDate = null;
+        @JsonProperty("shared_doc")
+private Boolean sharedDoc = null;
+        @JsonProperty("hyperlinks_changed")
+private Boolean hyperlinksChanged = null;
+
+        /** Sets the application field. */
+        @JsonProperty("application")
+        public Builder withApplication(final @Nullable String value) {
+            this.application = value;
+            return this;
+        }
+
+        /** Sets the appVersion field. */
+        @JsonProperty("app_version")
+        public Builder withAppVersion(final @Nullable String value) {
+            this.appVersion = value;
+            return this;
+        }
+
+        /** Sets the template field. */
+        @JsonProperty("template")
+        public Builder withTemplate(final @Nullable String value) {
+            this.template = value;
+            return this;
+        }
+
+        /** Sets the totalTime field. */
+        @JsonProperty("total_time")
+        public Builder withTotalTime(final @Nullable int value) {
+            this.totalTime = value;
+            return this;
+        }
+
+        /** Sets the pages field. */
+        @JsonProperty("pages")
+        public Builder withPages(final @Nullable int value) {
+            this.pages = value;
+            return this;
+        }
+
+        /** Sets the words field. */
+        @JsonProperty("words")
+        public Builder withWords(final @Nullable int value) {
+            this.words = value;
+            return this;
+        }
+
+        /** Sets the characters field. */
+        @JsonProperty("characters")
+        public Builder withCharacters(final @Nullable int value) {
+            this.characters = value;
+            return this;
+        }
+
+        /** Sets the charactersWithSpaces field. */
+        @JsonProperty("characters_with_spaces")
+        public Builder withCharactersWithSpaces(final @Nullable int value) {
+            this.charactersWithSpaces = value;
+            return this;
+        }
+
+        /** Sets the lines field. */
+        @JsonProperty("lines")
+        public Builder withLines(final @Nullable int value) {
+            this.lines = value;
+            return this;
+        }
+
+        /** Sets the paragraphs field. */
+        @JsonProperty("paragraphs")
+        public Builder withParagraphs(final @Nullable int value) {
+            this.paragraphs = value;
+            return this;
+        }
+
+        /** Sets the company field. */
+        @JsonProperty("company")
+        public Builder withCompany(final @Nullable String value) {
+            this.company = value;
+            return this;
+        }
+
+        /** Sets the docSecurity field. */
+        @JsonProperty("doc_security")
+        public Builder withDocSecurity(final @Nullable int value) {
+            this.docSecurity = value;
+            return this;
+        }
+
+        /** Sets the scaleCrop field. */
+        @JsonProperty("scale_crop")
+        public Builder withScaleCrop(final @Nullable boolean value) {
+            this.scaleCrop = value;
+            return this;
+        }
+
+        /** Sets the linksUpToDate field. */
+        @JsonProperty("links_up_to_date")
+        public Builder withLinksUpToDate(final @Nullable boolean value) {
+            this.linksUpToDate = value;
+            return this;
+        }
+
+        /** Sets the sharedDoc field. */
+        @JsonProperty("shared_doc")
+        public Builder withSharedDoc(final @Nullable boolean value) {
+            this.sharedDoc = value;
+            return this;
+        }
+
+        /** Sets the hyperlinksChanged field. */
+        @JsonProperty("hyperlinks_changed")
+        public Builder withHyperlinksChanged(final @Nullable boolean value) {
+            this.hyperlinksChanged = value;
+            return this;
+        }
+
+        /** Builds the DocxAppProperties instance. */
+        public DocxAppProperties build() {
+            return new DocxAppProperties(
+                application,
+                appVersion,
+                template,
+                totalTime,
+                pages,
+                words,
+                characters,
+                charactersWithSpaces,
+                lines,
+                paragraphs,
+                company,
+                docSecurity,
+                scaleCrop,
+                linksUpToDate,
+                sharedDoc,
+                hyperlinksChanged
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/DocxMetadata.java

@@ -0,0 +1,93 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.Map;
+import com.fasterxml.jackson.databind.JsonNode;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Word document metadata.
+ *
+ * Extracted from DOCX files using shared Office Open XML metadata extraction.
+ * Integrates with {@code office_metadata} module for core/app/custom properties.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = DocxMetadata.Builder.class)
+public record DocxMetadata(
+    /**
+     * Core properties from docProps/core.xml (Dublin Core metadata)
+     *
+     * Contains title, creator, subject, keywords, dates, etc.
+     * Shared format across DOCX/PPTX/XLSX documents.
+     */
+    @Nullable @JsonProperty("core_properties") CoreProperties coreProperties,
+    /**
+     * Application properties from docProps/app.xml (Word-specific statistics)
+     *
+     * Contains word count, page count, paragraph count, editing time, etc.
+     * DOCX-specific variant of Office application properties.
+     */
+    @Nullable @JsonProperty("app_properties") DocxAppProperties appProperties,
+    /**
+     * Custom properties from docProps/custom.xml (user-defined properties)
+     *
+     * Contains key-value pairs defined by users or applications.
+     * Values can be strings, numbers, booleans, or dates.
+     */
+    @Nullable @JsonProperty("custom_properties") Map<String, JsonNode> customProperties
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("core_properties")
+private CoreProperties coreProperties = null;
+        @JsonProperty("app_properties")
+private DocxAppProperties appProperties = null;
+        @JsonProperty("custom_properties")
+private Map<String, JsonNode> customProperties = null;
+
+        /** Sets the coreProperties field. */
+        @JsonProperty("core_properties")
+        public Builder withCoreProperties(final @Nullable CoreProperties value) {
+            this.coreProperties = value;
+            return this;
+        }
+
+        /** Sets the appProperties field. */
+        @JsonProperty("app_properties")
+        public Builder withAppProperties(final @Nullable DocxAppProperties value) {
+            this.appProperties = value;
+            return this;
+        }
+
+        /** Sets the customProperties field. */
+        @JsonProperty("custom_properties")
+        public Builder withCustomProperties(final @Nullable Map<String, JsonNode> value) {
+            this.customProperties = value;
+            return this;
+        }
+
+        /** Builds the DocxMetadata instance. */
+        public DocxMetadata build() {
+            return new DocxMetadata(
+                coreProperties,
+                appProperties,
+                customProperties
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/Element.java

@@ -0,0 +1,93 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * Semantic element extracted from document.
+ *
+ * Represents a logical unit of content with semantic classification,
+ * unique identifier, and metadata for tracking origin and position.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = Element.Builder.class)
+public record Element(
+    /**
+     * Unique element identifier
+     */
+    @JsonProperty("element_id") String elementId,
+    /**
+     * Semantic type of this element
+     */
+    @JsonProperty("element_type") ElementType elementType,
+    /**
+     * Text content of the element
+     */
+    @JsonProperty("text") String text,
+    /**
+     * Metadata about the element
+     */
+    @JsonProperty("metadata") ElementMetadata metadata
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("element_id")
+private String elementId = "";
+        @JsonProperty("element_type")
+private ElementType elementType = null;
+private String text = "";
+private ElementMetadata metadata = null;
+
+        /** Sets the elementId field. */
+        @JsonProperty("element_id")
+        public Builder withElementId(final String value) {
+            this.elementId = value;
+            return this;
+        }
+
+        /** Sets the elementType field. */
+        @JsonProperty("element_type")
+        public Builder withElementType(final ElementType value) {
+            this.elementType = value;
+            return this;
+        }
+
+        /** Sets the text field. */
+        @JsonProperty("text")
+        public Builder withText(final String value) {
+            this.text = value;
+            return this;
+        }
+
+        /** Sets the metadata field. */
+        @JsonProperty("metadata")
+        public Builder withMetadata(final ElementMetadata value) {
+            this.metadata = value;
+            return this;
+        }
+
+        /** Builds the Element instance. */
+        public Element build() {
+            return new Element(
+                elementId,
+                elementType,
+                text,
+                metadata
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/ElementMetadata.java

@@ -0,0 +1,105 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.Map;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Metadata for a semantic element.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ElementMetadata.Builder.class)
+public record ElementMetadata(
+    /**
+     * Page number (1-indexed)
+     */
+    @Nullable @JsonProperty("page_number") Integer pageNumber,
+    /**
+     * Source filename or document name
+     */
+    @Nullable @JsonProperty("filename") String filename,
+    /**
+     * Bounding box coordinates if available
+     */
+    @Nullable @JsonProperty("coordinates") BoundingBox coordinates,
+    /**
+     * Position index in the element sequence
+     */
+    @Nullable @JsonProperty("element_index") Long elementIndex,
+    /**
+     * Additional custom metadata
+     */
+    @JsonProperty("additional") Map<String, String> additional
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("page_number")
+private Integer pageNumber = null;
+private String filename = null;
+private BoundingBox coordinates = null;
+        @JsonProperty("element_index")
+private Long elementIndex = null;
+private Map<String, String> additional = Map.of();
+
+        /** Sets the pageNumber field. */
+        @JsonProperty("page_number")
+        public Builder withPageNumber(final @Nullable int value) {
+            this.pageNumber = value;
+            return this;
+        }
+
+        /** Sets the filename field. */
+        @JsonProperty("filename")
+        public Builder withFilename(final @Nullable String value) {
+            this.filename = value;
+            return this;
+        }
+
+        /** Sets the coordinates field. */
+        @JsonProperty("coordinates")
+        public Builder withCoordinates(final @Nullable BoundingBox value) {
+            this.coordinates = value;
+            return this;
+        }
+
+        /** Sets the elementIndex field. */
+        @JsonProperty("element_index")
+        public Builder withElementIndex(final @Nullable long value) {
+            this.elementIndex = value;
+            return this;
+        }
+
+        /** Sets the additional field. */
+        @JsonProperty("additional")
+        public Builder withAdditional(final Map<String, String> value) {
+            this.additional = value;
+            return this;
+        }
+
+        /** Builds the ElementMetadata instance. */
+        public ElementMetadata build() {
+            return new ElementMetadata(
+                pageNumber,
+                filename,
+                coordinates,
+                elementIndex,
+                additional
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/ElementType.java

@@ -0,0 +1,70 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonCreator;
+import com.fasterxml.jackson.annotation.JsonValue;
+
+/**
+ * Semantic element type classification.
+ *
+ * Categorizes text content into semantic units for downstream processing.
+ * Supports the element types commonly found in Unstructured documents.
+ */
+public enum ElementType {
+    /** Document title */
+    Title("title"),
+    /** Main narrative text body */
+    NarrativeText("narrative_text"),
+    /** Section heading */
+    Heading("heading"),
+    /** List item (bullet, numbered, etc.) */
+    ListItem("list_item"),
+    /** Table element */
+    Table("table"),
+    /** Image element */
+    Image("image"),
+    /** Page break marker */
+    PageBreak("page_break"),
+    /** Code block */
+    CodeBlock("code_block"),
+    /** Block quote */
+    BlockQuote("block_quote"),
+    /** Footer text */
+    Footer("footer"),
+    /** Header text */
+    Header("header");
+
+    /** The string value. */
+    private final String value;
+
+    ElementType(final String value) {
+        this.value = value;
+    }
+
+    /** Returns the string value. */
+    @JsonValue
+    public String getValue() {
+        return value;
+    }
+
+    /** Creates an instance from a string value. */
+    @JsonCreator
+    public static ElementType fromValue(final String value) {
+        for (ElementType e : values()) {
+            if (e.value.equalsIgnoreCase(value)) {
+                return e;
+            }
+        }
+        throw new IllegalArgumentException("Unknown value: " + value);
+    }
+
+    /** Returns the wire-format string value (matches JSON serialization). */
+    @Override
+    public String toString() {
+        return value;
+    }
+}

packages/java/dev/kreuzberg/EmailAttachment.java

@@ -0,0 +1,120 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Email attachment representation.
+ *
+ * Contains metadata and optionally the content of an email attachment.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = EmailAttachment.Builder.class)
+public record EmailAttachment(
+    /**
+     * Attachment name (from Content-Disposition header)
+     */
+    @Nullable @JsonProperty("name") String name,
+    /**
+     * Filename of the attachment
+     */
+    @Nullable @JsonProperty("filename") String filename,
+    /**
+     * MIME type of the attachment
+     */
+    @Nullable @JsonProperty("mime_type") String mimeType,
+    /**
+     * Size in bytes
+     */
+    @Nullable @JsonProperty("size") Long size,
+    /**
+     * Whether this attachment is an image
+     */
+    @JsonProperty("is_image") boolean isImage,
+    /**
+     * Attachment data (if extracted).
+     * Uses {@code bytes.Bytes} for cheap cloning of large buffers.
+     */
+    @Nullable @JsonProperty("data") byte[] data
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String name = null;
+private String filename = null;
+        @JsonProperty("mime_type")
+private String mimeType = null;
+private Long size = null;
+        @JsonProperty("is_image")
+private boolean isImage = false;
+private byte[] data = null;
+
+        /** Sets the name field. */
+        @JsonProperty("name")
+        public Builder withName(final @Nullable String value) {
+            this.name = value;
+            return this;
+        }
+
+        /** Sets the filename field. */
+        @JsonProperty("filename")
+        public Builder withFilename(final @Nullable String value) {
+            this.filename = value;
+            return this;
+        }
+
+        /** Sets the mimeType field. */
+        @JsonProperty("mime_type")
+        public Builder withMimeType(final @Nullable String value) {
+            this.mimeType = value;
+            return this;
+        }
+
+        /** Sets the size field. */
+        @JsonProperty("size")
+        public Builder withSize(final @Nullable long value) {
+            this.size = value;
+            return this;
+        }
+
+        /** Sets the isImage field. */
+        @JsonProperty("is_image")
+        public Builder withIsImage(final boolean value) {
+            this.isImage = value;
+            return this;
+        }
+
+        /** Sets the data field. */
+        @JsonProperty("data")
+        public Builder withData(final @Nullable byte[] value) {
+            this.data = value;
+            return this;
+        }
+
+        /** Builds the EmailAttachment instance. */
+        public EmailAttachment build() {
+            return new EmailAttachment(
+                name,
+                filename,
+                mimeType,
+                size,
+                isImage,
+                data
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/EmailConfig.java

@@ -0,0 +1,46 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Configuration for email extraction.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = EmailConfig.Builder.class)
+public record EmailConfig(@Nullable @JsonProperty("msg_fallback_codepage") Integer msgFallbackCodepage) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("msg_fallback_codepage")
+private Integer msgFallbackCodepage = null;
+
+        /** Sets the msgFallbackCodepage field. */
+        @JsonProperty("msg_fallback_codepage")
+        public Builder withMsgFallbackCodepage(final @Nullable int value) {
+            this.msgFallbackCodepage = value;
+            return this;
+        }
+
+        /** Builds the EmailConfig instance. */
+        public EmailConfig build() {
+            return new EmailConfig(
+                msgFallbackCodepage
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/EmailExtractionResult.java

@@ -0,0 +1,205 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import java.util.Map;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Email extraction result.
+ *
+ * Complete representation of an extracted email message (.eml or .msg)
+ * including headers, body content, and attachments.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = EmailExtractionResult.Builder.class)
+public record EmailExtractionResult(
+    /**
+     * Email subject line
+     */
+    @Nullable @JsonProperty("subject") String subject,
+    /**
+     * Sender email address
+     */
+    @Nullable @JsonProperty("from_email") String fromEmail,
+    /**
+     * Primary recipient email addresses
+     */
+    @JsonProperty("to_emails") List<String> toEmails,
+    /**
+     * CC recipient email addresses
+     */
+    @JsonProperty("cc_emails") List<String> ccEmails,
+    /**
+     * BCC recipient email addresses
+     */
+    @JsonProperty("bcc_emails") List<String> bccEmails,
+    /**
+     * Email date/timestamp
+     */
+    @Nullable @JsonProperty("date") String date,
+    /**
+     * Message-ID header value
+     */
+    @Nullable @JsonProperty("message_id") String messageId,
+    /**
+     * Plain text version of the email body
+     */
+    @Nullable @JsonProperty("plain_text") String plainText,
+    /**
+     * HTML version of the email body
+     */
+    @Nullable @JsonProperty("html_content") String htmlContent,
+    /**
+     * Cleaned/processed text content. Aliased as {@code cleaned_text} for back-compat.
+     */
+    @JsonProperty("content") String content,
+    /**
+     * List of email attachments
+     */
+    @JsonProperty("attachments") List<EmailAttachment> attachments,
+    /**
+     * Additional email headers and metadata
+     */
+    @JsonProperty("metadata") Map<String, String> metadata
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String subject = null;
+        @JsonProperty("from_email")
+private String fromEmail = null;
+        @JsonProperty("to_emails")
+private List<String> toEmails = List.of();
+        @JsonProperty("cc_emails")
+private List<String> ccEmails = List.of();
+        @JsonProperty("bcc_emails")
+private List<String> bccEmails = List.of();
+private String date = null;
+        @JsonProperty("message_id")
+private String messageId = null;
+        @JsonProperty("plain_text")
+private String plainText = null;
+        @JsonProperty("html_content")
+private String htmlContent = null;
+private String content = "";
+private List<EmailAttachment> attachments = List.of();
+private Map<String, String> metadata = Map.of();
+
+        /** Sets the subject field. */
+        @JsonProperty("subject")
+        public Builder withSubject(final @Nullable String value) {
+            this.subject = value;
+            return this;
+        }
+
+        /** Sets the fromEmail field. */
+        @JsonProperty("from_email")
+        public Builder withFromEmail(final @Nullable String value) {
+            this.fromEmail = value;
+            return this;
+        }
+
+        /** Sets the toEmails field. */
+        @JsonProperty("to_emails")
+        public Builder withToEmails(final List<String> value) {
+            this.toEmails = value;
+            return this;
+        }
+
+        /** Sets the ccEmails field. */
+        @JsonProperty("cc_emails")
+        public Builder withCcEmails(final List<String> value) {
+            this.ccEmails = value;
+            return this;
+        }
+
+        /** Sets the bccEmails field. */
+        @JsonProperty("bcc_emails")
+        public Builder withBccEmails(final List<String> value) {
+            this.bccEmails = value;
+            return this;
+        }
+
+        /** Sets the date field. */
+        @JsonProperty("date")
+        public Builder withDate(final @Nullable String value) {
+            this.date = value;
+            return this;
+        }
+
+        /** Sets the messageId field. */
+        @JsonProperty("message_id")
+        public Builder withMessageId(final @Nullable String value) {
+            this.messageId = value;
+            return this;
+        }
+
+        /** Sets the plainText field. */
+        @JsonProperty("plain_text")
+        public Builder withPlainText(final @Nullable String value) {
+            this.plainText = value;
+            return this;
+        }
+
+        /** Sets the htmlContent field. */
+        @JsonProperty("html_content")
+        public Builder withHtmlContent(final @Nullable String value) {
+            this.htmlContent = value;
+            return this;
+        }
+
+        /** Sets the content field. */
+        @JsonProperty("content")
+        public Builder withContent(final String value) {
+            this.content = value;
+            return this;
+        }
+
+        /** Sets the attachments field. */
+        @JsonProperty("attachments")
+        public Builder withAttachments(final List<EmailAttachment> value) {
+            this.attachments = value;
+            return this;
+        }
+
+        /** Sets the metadata field. */
+        @JsonProperty("metadata")
+        public Builder withMetadata(final Map<String, String> value) {
+            this.metadata = value;
+            return this;
+        }
+
+        /** Builds the EmailExtractionResult instance. */
+        public EmailExtractionResult build() {
+            return new EmailExtractionResult(
+                subject,
+                fromEmail,
+                toEmails,
+                ccEmails,
+                bccEmails,
+                date,
+                messageId,
+                plainText,
+                htmlContent,
+                content,
+                attachments,
+                metadata
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/EmailMetadata.java

@@ -0,0 +1,137 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Email metadata extracted from .eml and .msg files.
+ *
+ * Includes sender/recipient information, message ID, and attachment list.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = EmailMetadata.Builder.class)
+public record EmailMetadata(
+    /**
+     * Sender's email address
+     */
+    @Nullable @JsonProperty("from_email") String fromEmail,
+    /**
+     * Sender's display name
+     */
+    @Nullable @JsonProperty("from_name") String fromName,
+    /**
+     * Primary recipients
+     */
+    @JsonProperty("to_emails") List<String> toEmails,
+    /**
+     * CC recipients
+     */
+    @JsonProperty("cc_emails") List<String> ccEmails,
+    /**
+     * BCC recipients
+     */
+    @JsonProperty("bcc_emails") List<String> bccEmails,
+    /**
+     * Message-ID header value
+     */
+    @Nullable @JsonProperty("message_id") String messageId,
+    /**
+     * List of attachment filenames
+     */
+    @JsonProperty("attachments") List<String> attachments
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("from_email")
+private String fromEmail = null;
+        @JsonProperty("from_name")
+private String fromName = null;
+        @JsonProperty("to_emails")
+private List<String> toEmails = List.of();
+        @JsonProperty("cc_emails")
+private List<String> ccEmails = List.of();
+        @JsonProperty("bcc_emails")
+private List<String> bccEmails = List.of();
+        @JsonProperty("message_id")
+private String messageId = null;
+private List<String> attachments = List.of();
+
+        /** Sets the fromEmail field. */
+        @JsonProperty("from_email")
+        public Builder withFromEmail(final @Nullable String value) {
+            this.fromEmail = value;
+            return this;
+        }
+
+        /** Sets the fromName field. */
+        @JsonProperty("from_name")
+        public Builder withFromName(final @Nullable String value) {
+            this.fromName = value;
+            return this;
+        }
+
+        /** Sets the toEmails field. */
+        @JsonProperty("to_emails")
+        public Builder withToEmails(final List<String> value) {
+            this.toEmails = value;
+            return this;
+        }
+
+        /** Sets the ccEmails field. */
+        @JsonProperty("cc_emails")
+        public Builder withCcEmails(final List<String> value) {
+            this.ccEmails = value;
+            return this;
+        }
+
+        /** Sets the bccEmails field. */
+        @JsonProperty("bcc_emails")
+        public Builder withBccEmails(final List<String> value) {
+            this.bccEmails = value;
+            return this;
+        }
+
+        /** Sets the messageId field. */
+        @JsonProperty("message_id")
+        public Builder withMessageId(final @Nullable String value) {
+            this.messageId = value;
+            return this;
+        }
+
+        /** Sets the attachments field. */
+        @JsonProperty("attachments")
+        public Builder withAttachments(final List<String> value) {
+            this.attachments = value;
+            return this;
+        }
+
+        /** Builds the EmailMetadata instance. */
+        public EmailMetadata build() {
+            return new EmailMetadata(
+                fromEmail,
+                fromName,
+                toEmails,
+                ccEmails,
+                bccEmails,
+                messageId,
+                attachments
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/EmbeddedChanges.java

@@ -0,0 +1,78 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * Changes to embedded archive children between two results.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = EmbeddedChanges.Builder.class)
+public record EmbeddedChanges(
+    /**
+     * Children present in {@code b} but not in {@code a} (matched by {@code path}).
+     */
+    @JsonProperty("added") List<ArchiveEntry> added,
+    /**
+     * Children present in {@code a} but not in {@code b} (matched by {@code path}).
+     */
+    @JsonProperty("removed") List<ArchiveEntry> removed,
+    /**
+     * Children present in both but with differing content (matched by {@code path}).
+     *
+     * Each entry holds the diff of the nested {@code ExtractionResult}.
+     */
+    @JsonProperty("changed") List<EmbeddedDiff> changed
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private List<ArchiveEntry> added = List.of();
+private List<ArchiveEntry> removed = List.of();
+private List<EmbeddedDiff> changed = List.of();
+
+        /** Sets the added field. */
+        @JsonProperty("added")
+        public Builder withAdded(final List<ArchiveEntry> value) {
+            this.added = value;
+            return this;
+        }
+
+        /** Sets the removed field. */
+        @JsonProperty("removed")
+        public Builder withRemoved(final List<ArchiveEntry> value) {
+            this.removed = value;
+            return this;
+        }
+
+        /** Sets the changed field. */
+        @JsonProperty("changed")
+        public Builder withChanged(final List<EmbeddedDiff> value) {
+            this.changed = value;
+            return this;
+        }
+
+        /** Builds the EmbeddedChanges instance. */
+        public EmbeddedChanges build() {
+            return new EmbeddedChanges(
+                added,
+                removed,
+                changed
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/EmbeddedDiff.java

@@ -0,0 +1,62 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * Diff for a single embedded archive entry that appears in both results.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = EmbeddedDiff.Builder.class)
+public record EmbeddedDiff(
+    /**
+     * Archive-relative path identifying this entry.
+     */
+    @JsonProperty("path") String path,
+    /**
+     * The recursive diff of the entry's extraction result.
+     */
+    @JsonProperty("diff") ExtractionDiff diff
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String path = "";
+private ExtractionDiff diff = null;
+
+        /** Sets the path field. */
+        @JsonProperty("path")
+        public Builder withPath(final String value) {
+            this.path = value;
+            return this;
+        }
+
+        /** Sets the diff field. */
+        @JsonProperty("diff")
+        public Builder withDiff(final ExtractionDiff value) {
+            this.diff = value;
+            return this;
+        }
+
+        /** Builds the EmbeddedDiff instance. */
+        public EmbeddedDiff build() {
+            return new EmbeddedDiff(
+                path,
+                diff
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/EmbeddedFile.java

@@ -0,0 +1,95 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonSerialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Embedded file descriptor extracted from the PDF name tree.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = EmbeddedFile.Builder.class)
+public record EmbeddedFile(
+    /**
+     * The filename as stored in the PDF name tree.
+     */
+    @JsonProperty("name") String name,
+    /**
+     * Raw file bytes from the embedded stream (already decompressed by lopdf).
+     */
+    @JsonSerialize(using = ByteArrayToIntArraySerializer.class) @JsonProperty("data") byte[] data,
+    /**
+     * Compressed byte count of the original stream (before decompression).
+     *
+     * Used by callers to compute the decompression ratio and detect zip-bomb-style
+     * attacks that embed a tiny compressed stream expanding to gigabytes of data.
+     */
+    @JsonProperty("compressed_size") long compressedSize,
+    /**
+     * MIME type if specified in the filespec, otherwise {@code None}.
+     */
+    @Nullable @JsonProperty("mime_type") String mimeType
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String name = "";
+private byte[] data = new byte[0];
+        @JsonProperty("compressed_size")
+private long compressedSize = 0;
+        @JsonProperty("mime_type")
+private String mimeType = null;
+
+        /** Sets the name field. */
+        @JsonProperty("name")
+        public Builder withName(final String value) {
+            this.name = value;
+            return this;
+        }
+
+        /** Sets the data field. */
+        @JsonProperty("data")
+        public Builder withData(final byte[] value) {
+            this.data = value;
+            return this;
+        }
+
+        /** Sets the compressedSize field. */
+        @JsonProperty("compressed_size")
+        public Builder withCompressedSize(final long value) {
+            this.compressedSize = value;
+            return this;
+        }
+
+        /** Sets the mimeType field. */
+        @JsonProperty("mime_type")
+        public Builder withMimeType(final @Nullable String value) {
+            this.mimeType = value;
+            return this;
+        }
+
+        /** Builds the EmbeddedFile instance. */
+        public EmbeddedFile build() {
+            return new EmbeddedFile(
+                name,
+                data,
+                compressedSize,
+                mimeType
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/EmbeddingBackendBridge.java

@@ -0,0 +1,273 @@
+package dev.kreuzberg;
+
+import java.lang.foreign.Arena;
+import java.lang.foreign.FunctionDescriptor;
+import java.lang.foreign.Linker;
+import java.lang.foreign.MemoryLayout;
+import java.lang.foreign.MemorySegment;
+import java.lang.foreign.ValueLayout;
+import java.lang.invoke.MethodHandles;
+import java.lang.invoke.MethodType;
+import java.util.List;
+import java.util.concurrent.ConcurrentHashMap;
+import com.fasterxml.jackson.databind.ObjectMapper;
+
+/**
+ * Allocates Panama FFM upcall stubs for an IEmbeddingBackend implementation,
+ * assembles the C vtable in native memory, and provides static
+ * registerEmbeddingBackend/unregisterEmbeddingBackend helpers.
+ */
+public final class EmbeddingBackendBridge implements AutoCloseable {
+
+    private static final Linker LINKER = Linker.nativeLinker();
+    private static final MethodHandles.Lookup LOOKUP = MethodHandles.lookup();
+    private static final ObjectMapper JSON = new ObjectMapper();
+
+    /** Live registry — keeps Arenas and upcall stubs alive past the register call. */
+    private static final ConcurrentHashMap<String, EmbeddingBackendBridge>
+            EMBEDDING_BACKEND_BRIDGES = new ConcurrentHashMap<>();
+
+    // C vtable: 8 fields (4 plugin methods + 2 trait methods + free_string + free_user_data)
+    private static final MemoryLayout VTABLE_LAYOUT = MemoryLayout.structLayout(ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS);
+    private static final long VTABLE_SIZE = VTABLE_LAYOUT.byteSize();
+
+    private final Arena arena;
+    private final MemorySegment vtable;
+    private final IEmbeddingBackend impl;
+
+    EmbeddingBackendBridge(final IEmbeddingBackend impl) {
+        this.impl = impl;
+        this.arena = Arena.ofShared();
+        this.vtable = arena.allocate(VTABLE_SIZE);
+
+        try {
+            long offset = 0L;
+
+            var stubName = LINKER.upcallStub(LOOKUP.bind(this, "handleName",
+                MethodType.methodType(int.class, MemorySegment.class, MemorySegment.class, MemorySegment.class)),
+                FunctionDescriptor.of(ValueLayout.JAVA_INT, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubName);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+            var stubVersion = LINKER.upcallStub(LOOKUP.bind(this, "handleVersion",
+                MethodType.methodType(int.class, MemorySegment.class, MemorySegment.class, MemorySegment.class)),
+                FunctionDescriptor.of(ValueLayout.JAVA_INT, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubVersion);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+            var stubInitialize = LINKER.upcallStub(LOOKUP.bind(this, "handleInitialize",
+                MethodType.methodType(int.class, MemorySegment.class, MemorySegment.class)),
+                FunctionDescriptor.of(ValueLayout.JAVA_INT, ValueLayout.ADDRESS, ValueLayout.ADDRESS),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubInitialize);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+            var stubShutdown = LINKER.upcallStub(LOOKUP.bind(this, "handleShutdown",
+                MethodType.methodType(int.class, MemorySegment.class, MemorySegment.class)),
+                FunctionDescriptor.of(ValueLayout.JAVA_INT, ValueLayout.ADDRESS, ValueLayout.ADDRESS),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubShutdown);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+            var stubDimensions = LINKER.upcallStub(LOOKUP.bind(this, "handleDimensions",
+                MethodType.methodType(int.class, MemorySegment.class, MemorySegment.class, MemorySegment.class)),
+                FunctionDescriptor.of(ValueLayout.JAVA_INT, ValueLayout.ADDRESS, ValueLayout.ADDRESS, ValueLayout.ADDRESS),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubDimensions);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+            var stubEmbed = LINKER.upcallStub(LOOKUP.bind(this, "handleEmbed",
+                MethodType.methodType(int.class, MemorySegment.class, MemorySegment.class, MemorySegment.class, MemorySegment.class)),
+                FunctionDescriptor.of(
+                    ValueLayout.JAVA_INT,
+                    ValueLayout.ADDRESS,
+                    ValueLayout.ADDRESS,
+                    ValueLayout.ADDRESS,
+                    ValueLayout.ADDRESS
+                ),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubEmbed);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+            var stubFreeString = LINKER.upcallStub(LOOKUP.bind(this, "freeString",
+                MethodType.methodType(void.class, MemorySegment.class)),
+                FunctionDescriptor.ofVoid(ValueLayout.ADDRESS),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubFreeString);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+            var stubFreeUserData = LINKER.upcallStub(LOOKUP.bind(this, "freeUserData",
+                MethodType.methodType(void.class, MemorySegment.class)),
+                FunctionDescriptor.ofVoid(ValueLayout.ADDRESS),
+                arena);
+            vtable.set(ValueLayout.ADDRESS, offset, stubFreeUserData);
+            offset += ValueLayout.ADDRESS.byteSize();
+
+
+        } catch (ReflectiveOperationException e) {
+            arena.close();
+            throw new RuntimeException("Failed to create trait bridge stubs", e);
+        }
+    }
+
+    MemorySegment vtableSegment() { return vtable; }
+
+    private int handleName(MemorySegment userData, MemorySegment outName, MemorySegment outError) {
+        try {
+            outName.set(ValueLayout.ADDRESS, 0, arena.allocateFrom(impl.name()));
+            return 0;
+        } catch (Throwable e) { return 1; }
+    }
+
+    private int handleVersion(MemorySegment userData, MemorySegment outVersion, MemorySegment outError) {
+        try {
+            outVersion.set(ValueLayout.ADDRESS, 0, arena.allocateFrom(impl.version()));
+            return 0;
+        } catch (Throwable e) { return 1; }
+    }
+
+    private int handleInitialize(MemorySegment userData, MemorySegment outError) {
+        try {
+            impl.initialize();
+            return 0;
+        } catch (Throwable e) { return 1; }
+    }
+
+    private int handleShutdown(MemorySegment userData, MemorySegment outError) {
+        try {
+            impl.shutdown();
+            return 0;
+        } catch (Throwable e) { return 1; }
+    }
+
+    private int handleDimensions(MemorySegment userData, MemorySegment outResult, MemorySegment outError) {
+        try {
+            long result = impl.dimensions();
+            String json = JSON.writeValueAsString(result);
+            MemorySegment jsonCs = arena.allocateFrom(json);
+            outResult.set(ValueLayout.ADDRESS, 0, jsonCs);
+            return 0;
+        } catch (Throwable e) {
+            writeError(outError, e);
+            return 1;
+        }
+    }
+
+    private int handleEmbed(MemorySegment userData, MemorySegment texts_in, MemorySegment outResult, MemorySegment outError) {
+        try {
+            String texts_json = texts_in.reinterpret(Long.MAX_VALUE).getString(0);
+            List<String> texts = JSON.readValue(texts_json, new com.fasterxml.jackson.core.type.TypeReference<List<String>>() { });
+            List<List<Float>> result = impl.embed(texts);
+            String json = JSON.writeValueAsString(result);
+            MemorySegment jsonCs = arena.allocateFrom(json);
+            outResult.set(ValueLayout.ADDRESS, 0, jsonCs);
+            return 0;
+        } catch (Throwable e) {
+            writeError(outError, e);
+            return 1;
+        }
+    }
+
+    private void writeError(MemorySegment outError, Throwable e) {
+        try { outError.set(ValueLayout.ADDRESS, 0, arena.allocateFrom(e.getClass().getSimpleName() + ": " + e.getMessage())); }
+        catch (Throwable ignored) { /* swallow */ }
+    }
+
+    private void freeString(MemorySegment ptr) {
+        // Strings returned by Java callbacks are arena-owned and released when this bridge closes.
+    }
+
+    private void freeUserData(MemorySegment userData) {
+        // User data is Java-side state (the impl object), not freed by Rust on drop.
+    }
+
+    /** Read a NUL-terminated native C string safely without unbounded reinterpret. */
+    private static String readNativeString(MemorySegment ptr) {
+        return ptr.reinterpret(4096).getString(0);
+    }
+
+    @Override
+    public void close() { arena.close(); }
+
+    /** Register a EmbeddingBackend implementation via Panama FFM upcall stubs. */
+    public static void registerEmbeddingBackend(final IEmbeddingBackend impl) throws Exception {
+        var bridge = new EmbeddingBackendBridge(impl);
+        try {
+            try (var nameArena = Arena.ofShared()) {
+                var nameCs = nameArena.allocateFrom(impl.name());
+                MemorySegment outErr = nameArena.allocate(ValueLayout.ADDRESS);
+                int rc = (int) NativeLib.KREUZBERG_REGISTER_EMBEDDING_BACKEND.invoke(
+                    nameCs,
+                    bridge.vtableSegment(),
+                    MemorySegment.NULL,
+                    outErr
+                );
+                if (rc != 0) {
+                    MemorySegment errPtr = outErr.get(ValueLayout.ADDRESS, 0);
+                    String msg = errPtr.equals(MemorySegment.NULL) ? "registration failed (rc=" + rc + ")" : readNativeString(errPtr);
+                    throw new RuntimeException("registerEmbeddingBackend: " + msg);
+                }
+            }
+        } catch (Throwable t) {
+            bridge.close();
+            if (t instanceof Exception e) {
+                throw e;
+            } else {
+                throw new RuntimeException("Unexpected error during registration", t);
+            }
+        }
+        EMBEDDING_BACKEND_BRIDGES.put(impl.name(), bridge);
+    }
+
+    /** Unregister a EmbeddingBackend implementation by name. */
+    public static void unregisterEmbeddingBackend(String name) throws Exception {
+        try {
+            try (var nameArena = Arena.ofShared()) {
+                var nameCs = nameArena.allocateFrom(name);
+                MemorySegment outErr = nameArena.allocate(ValueLayout.ADDRESS);
+                int rc = (int) NativeLib.KREUZBERG_UNREGISTER_EMBEDDING_BACKEND.invoke(nameCs, outErr);
+                if (rc != 0) {
+                    MemorySegment errPtr = outErr.get(ValueLayout.ADDRESS, 0);
+                    String msg = errPtr.equals(MemorySegment.NULL)
+                        ? "unregistration failed (rc=" + rc + ")"
+                        : errPtr.reinterpret(Long.MAX_VALUE).getString(0);
+                    throw new RuntimeException("unregisterEmbeddingBackend: " + msg);
+                }
+            }
+        } catch (Throwable t) {
+            if (t instanceof Exception e) {
+                throw e;
+            } else {
+                throw new RuntimeException("Unexpected error during unregistration", t);
+            }
+        }
+        EmbeddingBackendBridge old = EMBEDDING_BACKEND_BRIDGES.remove(name);
+        if (old != null) { old.close(); }
+    }
+    /** Clear all registered EmbeddingBackend implementations. */
+    public static void clearEmbeddingBackends() throws Exception {
+        try {
+            try (var arena = Arena.ofShared()) {
+                MemorySegment outErr = arena.allocate(ValueLayout.ADDRESS);
+                int rc = (int) NativeLib.KREUZBERG_CLEAR_EMBEDDING_BACKEND.invoke(outErr);
+                if (rc != 0) {
+                    MemorySegment errPtr = outErr.get(ValueLayout.ADDRESS, 0);
+                    String msg = errPtr.equals(MemorySegment.NULL)
+                        ? "clear failed (rc=" + rc + ")"
+                        : errPtr.reinterpret(Long.MAX_VALUE).getString(0);
+                    throw new RuntimeException("clearEmbeddingBackends: " + msg);
+                }
+            }
+        } catch (Throwable t) {
+            if (t instanceof Exception e) {
+                throw e;
+            } else {
+                throw new RuntimeException("Unexpected error during clear", t);
+            }
+        }
+        EMBEDDING_BACKEND_BRIDGES.values().forEach(EmbeddingBackendBridge::close);
+        EMBEDDING_BACKEND_BRIDGES.clear();
+    }
+}

packages/java/dev/kreuzberg/EmbeddingConfig.java

@@ -0,0 +1,157 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Embedding configuration for text chunks.
+ *
+ * Configures embedding generation using ONNX models via the vendored embedding engine.
+ * Requires the {@code embeddings} feature to be enabled.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = EmbeddingConfig.Builder.class)
+public record EmbeddingConfig(
+    /**
+     * The embedding model to use (defaults to "balanced" preset if not specified)
+     */
+    @Nullable @JsonProperty("model") EmbeddingModelType model,
+    /**
+     * Whether to normalize embedding vectors (recommended for cosine similarity)
+     */
+    @Nullable @JsonProperty("normalize") Boolean normalize,
+    /**
+     * Batch size for embedding generation
+     */
+    @Nullable @JsonProperty("batch_size") Long batchSize,
+    /**
+     * Show model download progress
+     */
+    @Nullable @JsonProperty("show_download_progress") Boolean showDownloadProgress,
+    /**
+     * Custom cache directory for model files
+     *
+     * Defaults to {@code ~/.cache/kreuzberg/embeddings/} if not specified.
+     * Allows full customization of model download location.
+     */
+    @JsonProperty("cache_dir") java.nio.file.@Nullable Path cacheDir,
+    /**
+     * Hardware acceleration for the embedding ONNX model.
+     *
+     * When set, controls which execution provider (CPU, CUDA, CoreML, TensorRT)
+     * is used for inference. Defaults to {@code None} (auto-select per platform).
+     */
+    @Nullable @JsonProperty("acceleration") AccelerationConfig acceleration,
+    /**
+     * Maximum wall-clock duration (in seconds) for a single {@code embed()} call when
+     * using EmbeddingModelType.Plugin.
+     *
+     * Applies only to the in-process plugin path — protects against hung
+     * host-language backends (e.g. a Python callback deadlocked on the GIL,
+     * a model stuck on CUDA OOM retries, etc.). On timeout, the dispatcher
+     * returns {@code Plugin} instead of blocking forever.
+     *
+     * {@code None} disables the timeout. The default (60 seconds) is conservative
+     * for common in-process inference; increase for large batches on slow
+     * hardware.
+     */
+    @Nullable @JsonProperty("max_embed_duration_secs") Long maxEmbedDurationSecs
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+    public EmbeddingConfig{
+        if (batchSize == null) batchSize = 32L;
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @Nullable private EmbeddingModelType model = null;
+private Boolean normalize = null;
+        @JsonProperty("batch_size")
+private Long batchSize = null;
+        @JsonProperty("show_download_progress")
+private Boolean showDownloadProgress = null;
+        @JsonProperty("cache_dir")
+private java.nio.file.Path cacheDir = null;
+        @Nullable private AccelerationConfig acceleration = null;
+        @JsonProperty("max_embed_duration_secs")
+private Long maxEmbedDurationSecs = null;
+
+        /** Sets the model field. */
+        @JsonProperty("model")
+        public Builder withModel(final @Nullable EmbeddingModelType value) {
+            this.model = value;
+            return this;
+        }
+
+        /** Sets the normalize field. */
+        @JsonProperty("normalize")
+        public Builder withNormalize(final @Nullable Boolean value) {
+            this.normalize = value;
+            return this;
+        }
+
+        /** Sets the batchSize field. */
+        @JsonProperty("batch_size")
+        public Builder withBatchSize(final @Nullable Long value) {
+            this.batchSize = value;
+            return this;
+        }
+
+        /** Sets the showDownloadProgress field. */
+        @JsonProperty("show_download_progress")
+        public Builder withShowDownloadProgress(final @Nullable Boolean value) {
+            this.showDownloadProgress = value;
+            return this;
+        }
+
+        /** Sets the cacheDir field. */
+        @JsonProperty("cache_dir")
+        public Builder withCacheDir(final java.nio.file.@Nullable Path value) {
+            this.cacheDir = value;
+            return this;
+        }
+
+        /** Sets the acceleration field. */
+        @JsonProperty("acceleration")
+        public Builder withAcceleration(final @Nullable AccelerationConfig value) {
+            this.acceleration = value;
+            return this;
+        }
+
+        /** Sets the maxEmbedDurationSecs field. */
+        @JsonProperty("max_embed_duration_secs")
+        public Builder withMaxEmbedDurationSecs(final @Nullable Long value) {
+            this.maxEmbedDurationSecs = value;
+            return this;
+        }
+
+        /** Builds the EmbeddingConfig instance. */
+        public EmbeddingConfig build() {
+            return new EmbeddingConfig(
+                model,
+                normalize,
+                batchSize,
+                showDownloadProgress,
+                cacheDir,
+                acceleration,
+                maxEmbedDurationSecs
+            );
+        }
+    }
+    // CPD-ON
+    public static EmbeddingConfig defaultInstance() {
+        throw new UnsupportedOperationException("defaultInstance is not yet bridged via JNI; use the Builder instead.");
+    }
+}

packages/java/dev/kreuzberg/EmbeddingException.java

@@ -0,0 +1,15 @@
+// DO NOT EDIT - auto-generated by alef
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+package dev.kreuzberg;
+
+public class EmbeddingException extends KreuzbergErrorException {
+    /** Creates a new EmbeddingException with the given message. */
+    public EmbeddingException(final String message) {
+        super(message);
+    }
+
+    /** Creates a new EmbeddingException with the given message and cause. */
+    public EmbeddingException(final String message, final Throwable cause) {
+        super(message, cause);
+    }
+}

packages/java/dev/kreuzberg/EmbeddingModelType.java

@@ -0,0 +1,40 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonSubTypes;
+import com.fasterxml.jackson.annotation.JsonTypeInfo;
+
+/**
+ * Embedding model types supported by Kreuzberg.
+ */
+@JsonTypeInfo(use = JsonTypeInfo.Id.NAME, property = "type", visible = false)
+@JsonSubTypes({
+    @JsonSubTypes.Type(value = EmbeddingModelType.Preset.class, name = "preset"),
+    @JsonSubTypes.Type(value = EmbeddingModelType.Custom.class, name = "custom"),
+    @JsonSubTypes.Type(value = EmbeddingModelType.Llm.class, name = "llm"),
+    @JsonSubTypes.Type(value = EmbeddingModelType.Plugin.class, name = "plugin")
+})
+@com.fasterxml.jackson.annotation.JsonIgnoreProperties(ignoreUnknown = true)
+public sealed interface EmbeddingModelType {
+
+    /** Use a preset model configuration (recommended) */
+    record Preset(@JsonProperty("name") String name) implements EmbeddingModelType { }
+
+    /** Use a custom ONNX model from HuggingFace */
+    record Custom(
+        @JsonProperty("model_id") String modelId,
+        @JsonProperty("dimensions") long dimensions
+    ) implements EmbeddingModelType {
+    }
+
+    /** Provider-hosted embedding model via liter-llm. */
+    record Llm(@JsonProperty("llm") LlmConfig llm) implements EmbeddingModelType { }
+
+    /** In-process embedding backend registered via the plugin system. */
+    record Plugin(@JsonProperty("name") String name) implements EmbeddingModelType { }
+}

packages/java/dev/kreuzberg/EmbeddingPreset.java

@@ -0,0 +1,134 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * Preset configurations for common RAG use cases.
+ *
+ * Each preset combines chunk size, overlap, and embedding model
+ * to provide an optimized configuration for specific scenarios.
+ *
+ * All string fields are owned {@code String} for FFI compatibility — instances
+ * are safe to clone and pass across language boundaries.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = EmbeddingPreset.Builder.class)
+public record EmbeddingPreset(
+    @JsonProperty("name") String name,
+    @JsonProperty("chunk_size") long chunkSize,
+    @JsonProperty("overlap") long overlap,
+    /**
+     * HuggingFace repository name for the model.
+     */
+    @JsonProperty("model_repo") String modelRepo,
+    /**
+     * Pooling strategy: "cls" or "mean".
+     */
+    @JsonProperty("pooling") String pooling,
+    /**
+     * Path to the ONNX model file within the repo.
+     */
+    @JsonProperty("model_file") String modelFile,
+    @JsonProperty("dimensions") long dimensions,
+    @JsonProperty("description") String description
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String name = "";
+        @JsonProperty("chunk_size")
+private long chunkSize = 0;
+private long overlap = 0;
+        @JsonProperty("model_repo")
+private String modelRepo = "";
+private String pooling = "";
+        @JsonProperty("model_file")
+private String modelFile = "";
+private long dimensions = 0;
+private String description = "";
+
+        /** Sets the name field. */
+        @JsonProperty("name")
+        public Builder withName(final String value) {
+            this.name = value;
+            return this;
+        }
+
+        /** Sets the chunkSize field. */
+        @JsonProperty("chunk_size")
+        public Builder withChunkSize(final long value) {
+            this.chunkSize = value;
+            return this;
+        }
+
+        /** Sets the overlap field. */
+        @JsonProperty("overlap")
+        public Builder withOverlap(final long value) {
+            this.overlap = value;
+            return this;
+        }
+
+        /** Sets the modelRepo field. */
+        @JsonProperty("model_repo")
+        public Builder withModelRepo(final String value) {
+            this.modelRepo = value;
+            return this;
+        }
+
+        /** Sets the pooling field. */
+        @JsonProperty("pooling")
+        public Builder withPooling(final String value) {
+            this.pooling = value;
+            return this;
+        }
+
+        /** Sets the modelFile field. */
+        @JsonProperty("model_file")
+        public Builder withModelFile(final String value) {
+            this.modelFile = value;
+            return this;
+        }
+
+        /** Sets the dimensions field. */
+        @JsonProperty("dimensions")
+        public Builder withDimensions(final long value) {
+            this.dimensions = value;
+            return this;
+        }
+
+        /** Sets the description field. */
+        @JsonProperty("description")
+        public Builder withDescription(final String value) {
+            this.description = value;
+            return this;
+        }
+
+        /** Builds the EmbeddingPreset instance. */
+        public EmbeddingPreset build() {
+            return new EmbeddingPreset(
+                name,
+                chunkSize,
+                overlap,
+                modelRepo,
+                pooling,
+                modelFile,
+                dimensions,
+                description
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/EpubMetadata.java

@@ -0,0 +1,100 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * EPUB metadata (Dublin Core extensions).
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = EpubMetadata.Builder.class)
+public record EpubMetadata(
+    @Nullable @JsonProperty("coverage") String coverage,
+    @Nullable @JsonProperty("dc_format") String dcFormat,
+    @Nullable @JsonProperty("relation") String relation,
+    @Nullable @JsonProperty("source") String source,
+    @Nullable @JsonProperty("dc_type") String dcType,
+    @Nullable @JsonProperty("cover_image") String coverImage
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String coverage = null;
+        @JsonProperty("dc_format")
+private String dcFormat = null;
+private String relation = null;
+private String source = null;
+        @JsonProperty("dc_type")
+private String dcType = null;
+        @JsonProperty("cover_image")
+private String coverImage = null;
+
+        /** Sets the coverage field. */
+        @JsonProperty("coverage")
+        public Builder withCoverage(final @Nullable String value) {
+            this.coverage = value;
+            return this;
+        }
+
+        /** Sets the dcFormat field. */
+        @JsonProperty("dc_format")
+        public Builder withDcFormat(final @Nullable String value) {
+            this.dcFormat = value;
+            return this;
+        }
+
+        /** Sets the relation field. */
+        @JsonProperty("relation")
+        public Builder withRelation(final @Nullable String value) {
+            this.relation = value;
+            return this;
+        }
+
+        /** Sets the source field. */
+        @JsonProperty("source")
+        public Builder withSource(final @Nullable String value) {
+            this.source = value;
+            return this;
+        }
+
+        /** Sets the dcType field. */
+        @JsonProperty("dc_type")
+        public Builder withDcType(final @Nullable String value) {
+            this.dcType = value;
+            return this;
+        }
+
+        /** Sets the coverImage field. */
+        @JsonProperty("cover_image")
+        public Builder withCoverImage(final @Nullable String value) {
+            this.coverImage = value;
+            return this;
+        }
+
+        /** Builds the EpubMetadata instance. */
+        public EpubMetadata build() {
+            return new EpubMetadata(
+                coverage,
+                dcFormat,
+                relation,
+                source,
+                dcType,
+                coverImage
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/ErrorMetadata.java

@@ -0,0 +1,57 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * Error metadata (for batch operations).
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ErrorMetadata.Builder.class)
+public record ErrorMetadata(
+    @JsonProperty("error_type") String errorType,
+    @JsonProperty("message") String message
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("error_type")
+private String errorType = "";
+private String message = "";
+
+        /** Sets the errorType field. */
+        @JsonProperty("error_type")
+        public Builder withErrorType(final String value) {
+            this.errorType = value;
+            return this;
+        }
+
+        /** Sets the message field. */
+        @JsonProperty("message")
+        public Builder withMessage(final String value) {
+            this.message = value;
+            return this;
+        }
+
+        /** Builds the ErrorMetadata instance. */
+        public ErrorMetadata build() {
+            return new ErrorMetadata(
+                errorType,
+                message
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/ExcelMetadata.java

@@ -0,0 +1,69 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Excel/spreadsheet format metadata.
+ *
+ * Identifies the document as a spreadsheet source via the {@code FormatMetadata.Excel}
+ * discriminant. Sheet count and sheet names are stored inside this struct.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ExcelMetadata.Builder.class)
+public record ExcelMetadata(
+    /**
+     * Number of sheets in the workbook.
+     */
+    @Nullable @JsonProperty("sheet_count") Integer sheetCount,
+    /**
+     * Names of all sheets in the workbook.
+     */
+    @Nullable @JsonProperty("sheet_names") List<String> sheetNames
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("sheet_count")
+private Integer sheetCount = null;
+        @JsonProperty("sheet_names")
+private List<String> sheetNames = null;
+
+        /** Sets the sheetCount field. */
+        @JsonProperty("sheet_count")
+        public Builder withSheetCount(final @Nullable int value) {
+            this.sheetCount = value;
+            return this;
+        }
+
+        /** Sets the sheetNames field. */
+        @JsonProperty("sheet_names")
+        public Builder withSheetNames(final @Nullable List<String> value) {
+            this.sheetNames = value;
+            return this;
+        }
+
+        /** Builds the ExcelMetadata instance. */
+        public ExcelMetadata build() {
+            return new ExcelMetadata(
+                sheetCount,
+                sheetNames
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/ExcelSheet.java

@@ -0,0 +1,125 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Single Excel worksheet.
+ *
+ * Represents one sheet from an Excel workbook with its content
+ * converted to Markdown format and dimensional statistics.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ExcelSheet.Builder.class)
+public record ExcelSheet(
+    /**
+     * Sheet name as it appears in Excel
+     */
+    @JsonProperty("name") String name,
+    /**
+     * Sheet content converted to Markdown tables
+     */
+    @JsonProperty("markdown") String markdown,
+    /**
+     * Number of rows
+     */
+    @JsonProperty("row_count") long rowCount,
+    /**
+     * Number of columns
+     */
+    @JsonProperty("col_count") long colCount,
+    /**
+     * Total number of non-empty cells
+     */
+    @JsonProperty("cell_count") long cellCount,
+    /**
+     * Pre-extracted table cells (2D vector of cell values)
+     * Populated during markdown generation to avoid re-parsing markdown.
+     * null for empty sheets.
+     */
+    @Nullable @JsonProperty("table_cells") List<List<String>> tableCells
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String name = "";
+private String markdown = "";
+        @JsonProperty("row_count")
+private long rowCount = 0;
+        @JsonProperty("col_count")
+private long colCount = 0;
+        @JsonProperty("cell_count")
+private long cellCount = 0;
+        @JsonProperty("table_cells")
+private List<List<String>> tableCells = null;
+
+        /** Sets the name field. */
+        @JsonProperty("name")
+        public Builder withName(final String value) {
+            this.name = value;
+            return this;
+        }
+
+        /** Sets the markdown field. */
+        @JsonProperty("markdown")
+        public Builder withMarkdown(final String value) {
+            this.markdown = value;
+            return this;
+        }
+
+        /** Sets the rowCount field. */
+        @JsonProperty("row_count")
+        public Builder withRowCount(final long value) {
+            this.rowCount = value;
+            return this;
+        }
+
+        /** Sets the colCount field. */
+        @JsonProperty("col_count")
+        public Builder withColCount(final long value) {
+            this.colCount = value;
+            return this;
+        }
+
+        /** Sets the cellCount field. */
+        @JsonProperty("cell_count")
+        public Builder withCellCount(final long value) {
+            this.cellCount = value;
+            return this;
+        }
+
+        /** Sets the tableCells field. */
+        @JsonProperty("table_cells")
+        public Builder withTableCells(final @Nullable List<List<String>> value) {
+            this.tableCells = value;
+            return this;
+        }
+
+        /** Builds the ExcelSheet instance. */
+        public ExcelSheet build() {
+            return new ExcelSheet(
+                name,
+                markdown,
+                rowCount,
+                colCount,
+                cellCount,
+                tableCells
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/ExcelWorkbook.java

@@ -0,0 +1,88 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import java.util.Map;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Excel workbook representation.
+ *
+ * Contains all sheets from an Excel file (.xlsx, .xls, etc.) with
+ * extracted content and metadata.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ExcelWorkbook.Builder.class)
+public record ExcelWorkbook(
+    /**
+     * All sheets in the workbook
+     */
+    @JsonProperty("sheets") List<ExcelSheet> sheets,
+    /**
+     * Workbook-level metadata (author, creation date, etc.)
+     */
+    @JsonProperty("metadata") Map<String, String> metadata,
+    /**
+     * Collaborative-edit revision headers from {@code xl/revisions/revisionHeaders.xml}.
+     *
+     * Populated for legacy shared-workbook {@code .xlsx} files that contain the
+     * {@code xl/revisions/} directory. Each {@code &lt;header&gt;} element maps to one
+     * {@code DocumentRevision { kind: FormatChange }} carrying the header's {@code guid}
+     * (→ {@code revision_id}), {@code userName} (→ {@code author}), and {@code dateTime} (→ {@code timestamp}).
+     * {@code anchor} and {@code delta} are {@code None}/empty for v1 (per-cell log parsing is a
+     * follow-up). {@code None} when {@code xl/revisions/revisionHeaders.xml} is absent.
+     */
+    @Nullable @JsonProperty("revisions") List<DocumentRevision> revisions
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private List<ExcelSheet> sheets = List.of();
+private Map<String, String> metadata = Map.of();
+private List<DocumentRevision> revisions = null;
+
+        /** Sets the sheets field. */
+        @JsonProperty("sheets")
+        public Builder withSheets(final List<ExcelSheet> value) {
+            this.sheets = value;
+            return this;
+        }
+
+        /** Sets the metadata field. */
+        @JsonProperty("metadata")
+        public Builder withMetadata(final Map<String, String> value) {
+            this.metadata = value;
+            return this;
+        }
+
+        /** Sets the revisions field. */
+        @JsonProperty("revisions")
+        public Builder withRevisions(final @Nullable List<DocumentRevision> value) {
+            this.revisions = value;
+            return this;
+        }
+
+        /** Builds the ExcelWorkbook instance. */
+        public ExcelWorkbook build() {
+            return new ExcelWorkbook(
+                sheets,
+                metadata,
+                revisions
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/ExecutionProviderType.java

@@ -0,0 +1,58 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonCreator;
+import com.fasterxml.jackson.annotation.JsonValue;
+
+/**
+ * ONNX Runtime execution provider type.
+ *
+ * Determines which hardware backend is used for model inference.
+ * {@code Auto} (default) selects the best available provider per platform.
+ */
+public enum ExecutionProviderType {
+    /** Auto-select: CoreML on macOS, CUDA on Linux, CPU elsewhere. */
+    Auto("auto"),
+    /** CPU execution provider (always available). */
+    Cpu("cpu"),
+    /** Apple CoreML (macOS/iOS Neural Engine + GPU). */
+    CoreMl("coreml"),
+    /** NVIDIA CUDA GPU acceleration. */
+    Cuda("cuda"),
+    /** NVIDIA TensorRT (optimized CUDA inference). */
+    TensorRt("tensorrt");
+
+    /** The string value. */
+    private final String value;
+
+    ExecutionProviderType(final String value) {
+        this.value = value;
+    }
+
+    /** Returns the string value. */
+    @JsonValue
+    public String getValue() {
+        return value;
+    }
+
+    /** Creates an instance from a string value. */
+    @JsonCreator
+    public static ExecutionProviderType fromValue(final String value) {
+        for (ExecutionProviderType e : values()) {
+            if (e.value.equalsIgnoreCase(value)) {
+                return e;
+            }
+        }
+        throw new IllegalArgumentException("Unknown value: " + value);
+    }
+
+    /** Returns the wire-format string value (matches JSON serialization). */
+    @Override
+    public String toString() {
+        return value;
+    }
+}

packages/java/dev/kreuzberg/ExtractedImage.java

@@ -0,0 +1,269 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonSerialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Extracted image from a document.
+ *
+ * Contains raw image data, metadata, and optional nested OCR results.
+ * Raw bytes allow cross-language compatibility - users can convert to
+ * PIL.Image (Python), Sharp (Node.js), or other formats as needed.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ExtractedImage.Builder.class)
+public record ExtractedImage(
+    /**
+     * Raw image data (PNG, JPEG, WebP, etc. bytes).
+     * Uses {@code bytes.Bytes} for cheap cloning of large buffers.
+     */
+    @JsonSerialize(using = ByteArrayToIntArraySerializer.class) @JsonProperty("data") byte[] data,
+    /**
+     * Image format (e.g., "jpeg", "png", "webp")
+     * Uses Cow<, str> to avoid allocation for static literals.
+     */
+    @JsonProperty("format") String format,
+    /**
+     * Zero-indexed position of this image in the document/page
+     */
+    @JsonProperty("image_index") int imageIndex,
+    /**
+     * Page/slide number where image was found (1-indexed)
+     */
+    @Nullable @JsonProperty("page_number") Integer pageNumber,
+    /**
+     * Image width in pixels
+     */
+    @Nullable @JsonProperty("width") Integer width,
+    /**
+     * Image height in pixels
+     */
+    @Nullable @JsonProperty("height") Integer height,
+    /**
+     * Colorspace information (e.g., "RGB", "CMYK", "Gray")
+     */
+    @Nullable @JsonProperty("colorspace") String colorspace,
+    /**
+     * Bits per color component (e.g., 8, 16)
+     */
+    @Nullable @JsonProperty("bits_per_component") Integer bitsPerComponent,
+    /**
+     * Whether this image is a mask image
+     */
+    @Nullable @JsonProperty("is_mask") Boolean isMask,
+    /**
+     * Optional description of the image
+     */
+    @Nullable @JsonProperty("description") String description,
+    /**
+     * Nested OCR extraction result (if image was OCRed)
+     *
+     * When OCR is performed on this image, the result is embedded here
+     * rather than in a separate collection, making the relationship explicit.
+     */
+    @Nullable @JsonProperty("ocr_result") ExtractionResult ocrResult,
+    /**
+     * Bounding box of the image on the page (PDF coordinates: x0=left, y0=bottom, x1=right, y1=top).
+     * Only populated for PDF-extracted images when position data is available from the PDF extractor.
+     */
+    @Nullable @JsonProperty("bounding_box") BoundingBox boundingBox,
+    /**
+     * Original source path of the image within the document archive (e.g., "media/image1.png" in DOCX).
+     * Used for rendering image references when the binary data is not extracted.
+     */
+    @Nullable @JsonProperty("source_path") String sourcePath,
+    /**
+     * Heuristic classification of what this image likely depicts.
+     * {@code None} if classification was disabled or inconclusive.
+     */
+    @Nullable @JsonProperty("image_kind") ImageKind imageKind,
+    /**
+     * Confidence score for {@code image_kind}, in the range 0.0 to 1.0.
+     */
+    @Nullable @JsonProperty("kind_confidence") Float kindConfidence,
+    /**
+     * Identifier shared across images that form a single logical figure
+     * (e.g. all raster tiles of one technical drawing). {@code None} for singletons.
+     */
+    @Nullable @JsonProperty("cluster_id") Integer clusterId
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private byte[] data = new byte[0];
+private String format = "";
+        @JsonProperty("image_index")
+private int imageIndex = 0;
+        @JsonProperty("page_number")
+private Integer pageNumber = null;
+private Integer width = null;
+private Integer height = null;
+private String colorspace = null;
+        @JsonProperty("bits_per_component")
+private Integer bitsPerComponent = null;
+        @JsonProperty("is_mask")
+private Boolean isMask = null;
+private String description = null;
+        @JsonProperty("ocr_result")
+private ExtractionResult ocrResult = null;
+        @JsonProperty("bounding_box")
+        @Nullable private BoundingBox boundingBox = null;
+        @JsonProperty("source_path")
+private String sourcePath = null;
+        @JsonProperty("image_kind")
+        @Nullable private ImageKind imageKind = null;
+        @JsonProperty("kind_confidence")
+private Float kindConfidence = null;
+        @JsonProperty("cluster_id")
+private Integer clusterId = null;
+
+        /** Sets the data field. */
+        @JsonProperty("data")
+        public Builder withData(final byte[] value) {
+            this.data = value;
+            return this;
+        }
+
+        /** Sets the format field. */
+        @JsonProperty("format")
+        public Builder withFormat(final String value) {
+            this.format = value;
+            return this;
+        }
+
+        /** Sets the imageIndex field. */
+        @JsonProperty("image_index")
+        public Builder withImageIndex(final int value) {
+            this.imageIndex = value;
+            return this;
+        }
+
+        /** Sets the pageNumber field. */
+        @JsonProperty("page_number")
+        public Builder withPageNumber(final @Nullable int value) {
+            this.pageNumber = value;
+            return this;
+        }
+
+        /** Sets the width field. */
+        @JsonProperty("width")
+        public Builder withWidth(final @Nullable int value) {
+            this.width = value;
+            return this;
+        }
+
+        /** Sets the height field. */
+        @JsonProperty("height")
+        public Builder withHeight(final @Nullable int value) {
+            this.height = value;
+            return this;
+        }
+
+        /** Sets the colorspace field. */
+        @JsonProperty("colorspace")
+        public Builder withColorspace(final @Nullable String value) {
+            this.colorspace = value;
+            return this;
+        }
+
+        /** Sets the bitsPerComponent field. */
+        @JsonProperty("bits_per_component")
+        public Builder withBitsPerComponent(final @Nullable int value) {
+            this.bitsPerComponent = value;
+            return this;
+        }
+
+        /** Sets the isMask field. */
+        @JsonProperty("is_mask")
+        public Builder withIsMask(final @Nullable Boolean value) {
+            this.isMask = value;
+            return this;
+        }
+
+        /** Sets the description field. */
+        @JsonProperty("description")
+        public Builder withDescription(final @Nullable String value) {
+            this.description = value;
+            return this;
+        }
+
+        /** Sets the ocrResult field. */
+        @JsonProperty("ocr_result")
+        public Builder withOcrResult(final @Nullable ExtractionResult value) {
+            this.ocrResult = value;
+            return this;
+        }
+
+        /** Sets the boundingBox field. */
+        @JsonProperty("bounding_box")
+        public Builder withBoundingBox(final @Nullable BoundingBox value) {
+            this.boundingBox = value;
+            return this;
+        }
+
+        /** Sets the sourcePath field. */
+        @JsonProperty("source_path")
+        public Builder withSourcePath(final @Nullable String value) {
+            this.sourcePath = value;
+            return this;
+        }
+
+        /** Sets the imageKind field. */
+        @JsonProperty("image_kind")
+        public Builder withImageKind(final @Nullable ImageKind value) {
+            this.imageKind = value;
+            return this;
+        }
+
+        /** Sets the kindConfidence field. */
+        @JsonProperty("kind_confidence")
+        public Builder withKindConfidence(final @Nullable Float value) {
+            this.kindConfidence = value;
+            return this;
+        }
+
+        /** Sets the clusterId field. */
+        @JsonProperty("cluster_id")
+        public Builder withClusterId(final @Nullable Integer value) {
+            this.clusterId = value;
+            return this;
+        }
+
+        /** Builds the ExtractedImage instance. */
+        public ExtractedImage build() {
+            return new ExtractedImage(
+                data,
+                format,
+                imageIndex,
+                pageNumber,
+                width,
+                height,
+                colorspace,
+                bitsPerComponent,
+                isMask,
+                description,
+                ocrResult,
+                boundingBox,
+                sourcePath,
+                imageKind,
+                kindConfidence,
+                clusterId
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/ExtractedUri.java

@@ -0,0 +1,93 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * A URI extracted from a document.
+ *
+ * Represents any link, reference, or resource pointer found during extraction.
+ * The {@code kind} field classifies the URI semantically, while {@code label} carries
+ * optional human-readable display text.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ExtractedUri.Builder.class)
+public record ExtractedUri(
+    /**
+     * The URL or path string.
+     */
+    @JsonProperty("url") String url,
+    /**
+     * Optional display text / label for the link.
+     */
+    @Nullable @JsonProperty("label") String label,
+    /**
+     * Optional page number where the URI was found (1-indexed).
+     */
+    @Nullable @JsonProperty("page") Integer page,
+    /**
+     * Semantic classification of the URI.
+     */
+    @JsonProperty("kind") UriKind kind
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String url = "";
+private String label = null;
+private Integer page = null;
+private UriKind kind = null;
+
+        /** Sets the url field. */
+        @JsonProperty("url")
+        public Builder withUrl(final String value) {
+            this.url = value;
+            return this;
+        }
+
+        /** Sets the label field. */
+        @JsonProperty("label")
+        public Builder withLabel(final @Nullable String value) {
+            this.label = value;
+            return this;
+        }
+
+        /** Sets the page field. */
+        @JsonProperty("page")
+        public Builder withPage(final @Nullable int value) {
+            this.page = value;
+            return this;
+        }
+
+        /** Sets the kind field. */
+        @JsonProperty("kind")
+        public Builder withKind(final UriKind value) {
+            this.kind = value;
+            return this;
+        }
+
+        /** Builds the ExtractedUri instance. */
+        public ExtractedUri build() {
+            return new ExtractedUri(
+                url,
+                label,
+                page,
+                kind
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/ExtractionConfig.java

@@ -0,0 +1,650 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Main extraction configuration.
+ *
+ * This struct contains all configuration options for the extraction process.
+ * It can be loaded from TOML, YAML, or JSON files, or created programmatically.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ExtractionConfig.Builder.class)
+public record ExtractionConfig(
+    /**
+     * Enable caching of extraction results
+     */
+    @Nullable @JsonProperty("use_cache") Boolean useCache,
+    /**
+     * Enable quality post-processing
+     */
+    @Nullable @JsonProperty("enable_quality_processing") Boolean enableQualityProcessing,
+    /**
+     * OCR configuration (null = OCR disabled)
+     */
+    @Nullable @JsonProperty("ocr") OcrConfig ocr,
+    /**
+     * Force OCR even for searchable PDFs
+     */
+    @Nullable @JsonProperty("force_ocr") Boolean forceOcr,
+    /**
+     * Force OCR on specific pages only (1-indexed page numbers, must be >= 1).
+     *
+     * When set, only the listed pages are OCR'd regardless of text layer quality.
+     * Unlisted pages use native text extraction. Ignored when {@code force_ocr} is {@code true}.
+     * Only applies to PDF documents. Duplicates are automatically deduplicated.
+     * An {@code ocr} config is recommended for backend/language selection; defaults are used if absent.
+     */
+    @Nullable @JsonProperty("force_ocr_pages") List<Integer> forceOcrPages,
+    /**
+     * Disable OCR entirely, even for images.
+     *
+     * When {@code true}, OCR is skipped for all document types. Images return metadata
+     * only (dimensions, format, EXIF) without text extraction. PDFs use only
+     * native text extraction without OCR fallback.
+     *
+     * Cannot be {@code true} simultaneously with {@code force_ocr}.
+     *
+     * *Added in v4.7.0.*
+     */
+    @Nullable @JsonProperty("disable_ocr") Boolean disableOcr,
+    /**
+     * Text chunking configuration (null = chunking disabled)
+     */
+    @Nullable @JsonProperty("chunking") ChunkingConfig chunking,
+    /**
+     * Content filtering configuration (null = use extractor defaults).
+     *
+     * Controls whether document "furniture" (headers, footers, watermarks,
+     * repeating text) is included in or stripped from extraction results.
+     * See ContentFilterConfig for per-field documentation.
+     */
+    @Nullable @JsonProperty("content_filter") ContentFilterConfig contentFilter,
+    /**
+     * Image extraction configuration (null = no image extraction)
+     */
+    @Nullable @JsonProperty("images") ImageExtractionConfig images,
+    /**
+     * PDF-specific options (null = use defaults)
+     */
+    @Nullable @JsonProperty("pdf_options") PdfConfig pdfOptions,
+    /**
+     * Token reduction configuration (null = no token reduction)
+     */
+    @Nullable @JsonProperty("token_reduction") TokenReductionOptions tokenReduction,
+    /**
+     * Language detection configuration (null = no language detection)
+     */
+    @Nullable @JsonProperty("language_detection") LanguageDetectionConfig languageDetection,
+    /**
+     * Page extraction configuration (null = no page tracking)
+     */
+    @Nullable @JsonProperty("pages") PageConfig pages,
+    /**
+     * Keyword extraction configuration (null = no keyword extraction)
+     */
+    @Nullable @JsonProperty("keywords") KeywordConfig keywords,
+    /**
+     * Post-processor configuration (null = use defaults)
+     */
+    @Nullable @JsonProperty("postprocessor") PostProcessorConfig postprocessor,
+    /**
+     * HTML to Markdown conversion options (null = use defaults)
+     *
+     * Configure how HTML documents are converted to Markdown, including heading styles,
+     * list formatting, code block styles, and preprocessing options.
+     */
+    @Nullable @JsonProperty("html_options") String htmlOptions,
+    /**
+     * Styled HTML output configuration.
+     *
+     * When set alongside {@code output_format = OutputFormat.Html}, the extraction
+     * pipeline uses StyledHtmlRenderer(crate.rendering.StyledHtmlRenderer)
+     * which emits stable {@code kb-*} CSS class hooks on every structural element
+     * and optionally embeds theme CSS or user-supplied CSS in a {@code &lt;style&gt;} block.
+     *
+     * When {@code None}, the existing plain comrak-based HTML renderer is used.
+     */
+    @Nullable @JsonProperty("html_output") HtmlOutputConfig htmlOutput,
+    /**
+     * Default per-file timeout in seconds for batch extraction.
+     *
+     * When set, each file in a batch will be canceled after this duration
+     * unless overridden by FileExtractionConfig.timeout_secs.
+     *
+     * Defaults to {@code Some(60)} to prevent pathological files (e.g. deeply
+     * nested archives, documents with millions of cells) from running
+     * indefinitely and exhausting caller resources. Set to {@code None} to
+     * disable the timeout for trusted input or long-running workloads.
+     */
+    @Nullable @JsonProperty("extraction_timeout_secs") Long extractionTimeoutSecs,
+    /**
+     * Maximum concurrent extractions in batch operations (null = (num_cpus × 1.5).ceil()).
+     *
+     * Limits parallelism to prevent resource exhaustion when processing
+     * large batches. Defaults to (num_cpus × 1.5).ceil() when not set.
+     */
+    @Nullable @JsonProperty("max_concurrent_extractions") Long maxConcurrentExtractions,
+    /**
+     * Result structure format
+     *
+     * Controls whether results are returned in unified format (default) with all
+     * content in the {@code content} field, or element-based format with semantic
+     * elements (for Unstructured-compatible output).
+     */
+    @Nullable @JsonProperty("result_format") ResultFormat resultFormat,
+    /**
+     * Security limits for archive extraction.
+     *
+     * Controls maximum archive size, compression ratio, file count, and other
+     * security thresholds to prevent decompression bomb attacks. Also caps
+     * nesting depth, iteration count, entity / token length, total
+     * content size, and table cell count for every extraction path that
+     * ingests user-controlled bytes.
+     * When {@code None}, default limits are used.
+     */
+    @Nullable @JsonProperty("security_limits") SecurityLimits securityLimits,
+    /**
+     * Maximum uncompressed size in bytes for a single embedded file before
+     * recursive extraction is attempted (default: 50 MiB).
+     *
+     * Applies to embedded objects inside OOXML containers (DOCX, PPTX) and
+     * to email attachments processed via recursive extraction. Files that
+     * exceed this limit are skipped with a {@code ProcessingWarning} rather than
+     * passed to the extraction pipeline, preventing a single oversized
+     * embedded object from consuming unbounded memory or time.
+     *
+     * Set to {@code None} to disable the per-embedded-file cap (falls back to
+     * {@code security_limits.max_archive_size} as the only guard).
+     */
+    @Nullable @JsonProperty("max_embedded_file_bytes") Long maxEmbeddedFileBytes,
+    /**
+     * Content text format (default: Plain).
+     *
+     * Controls the format of the extracted content:
+     * - {@code Plain}: Raw extracted text (default)
+     * - {@code Markdown}: Markdown formatted output
+     * - {@code Djot}: Djot markup format (requires djot feature)
+     * - {@code Html}: HTML formatted output
+     *
+     * When set to a structured format, extraction results will include
+     * formatted output. The {@code formatted_content} field may be populated
+     * when format conversion is applied.
+     */
+    @Nullable @JsonProperty("output_format") OutputFormat outputFormat,
+    /**
+     * Layout detection configuration (null = layout detection disabled).
+     *
+     * When set, PDF pages and images are analyzed for document structure
+     * (headings, code, formulas, tables, figures, etc.) using RT-DETR models
+     * via ONNX Runtime. For PDFs, layout hints override paragraph classification
+     * in the markdown pipeline. For images, per-region OCR is performed with
+     * markdown formatting based on detected layout classes.
+     * Requires the {@code layout-detection} feature to run inference; the field is
+     * present whenever the {@code layout-types} feature is active (which includes
+     * {@code layout-detection} as well as the no-ORT target groups).
+     */
+    @Nullable @JsonProperty("layout") LayoutDetectionConfig layout,
+    /**
+     * Run layout detection on the non-OCR PDF markdown path.
+     *
+     * When {@code true} and {@code layout} is {@code Some(_)}, layout regions inform heading,
+     * table, list, and figure detection in the structure pipeline that would
+     * otherwise rely on font-clustering heuristics alone. Significantly
+     * improves SF1 (structural F1) at the cost of inference latency
+     * (~150-300ms/page CPU, ~20-50ms/page GPU). Default: {@code false}.
+     * Requires the {@code layout-detection} feature.
+     */
+    @Nullable @JsonProperty("use_layout_for_markdown") Boolean useLayoutForMarkdown,
+    /**
+     * Enable structured document tree output.
+     *
+     * When true, populates the {@code document} field on {@code ExtractionResult} with a
+     * hierarchical {@code DocumentStructure} containing heading-driven section nesting,
+     * table grids, content layer classification, and inline annotations.
+     *
+     * Independent of {@code result_format} — can be combined with Unified or ElementBased.
+     */
+    @Nullable @JsonProperty("include_document_structure") Boolean includeDocumentStructure,
+    /**
+     * Hardware acceleration configuration for ONNX Runtime models.
+     *
+     * Controls execution provider selection for layout detection and embedding
+     * models. When {@code None}, uses platform defaults (CoreML on macOS, CUDA on
+     * Linux, CPU on Windows).
+     */
+    @Nullable @JsonProperty("acceleration") AccelerationConfig acceleration,
+    /**
+     * Cache namespace for tenant isolation.
+     *
+     * When set, cache entries are stored under {@code {cache_dir}/{namespace}/}.
+     * Must be alphanumeric, hyphens, or underscores only (max 64 chars).
+     * Different namespaces have isolated cache spaces on the same filesystem.
+     */
+    @Nullable @JsonProperty("cache_namespace") String cacheNamespace,
+    /**
+     * Per-request cache TTL in seconds.
+     *
+     * Overrides the global {@code max_age_days} for this specific extraction.
+     * When {@code 0}, caching is completely skipped (no read or write).
+     * When {@code None}, the global TTL applies.
+     */
+    @Nullable @JsonProperty("cache_ttl_secs") Long cacheTtlSecs,
+    /**
+     * Email extraction configuration (null = use defaults).
+     *
+     * Currently supports configuring the fallback codepage for MSG files
+     * that do not specify one. See {@code EmailConfig} for details.
+     */
+    @Nullable @JsonProperty("email") EmailConfig email,
+    /**
+     * Concurrency limits for constrained environments (null = use defaults).
+     *
+     * Controls Rayon thread pool size, ONNX Runtime intra-op threads, and
+     * (when {@code max_concurrent_extractions} is unset) the batch concurrency
+     * semaphore. See {@code ConcurrencyConfig} for details.
+     */
+    @Nullable @JsonProperty("concurrency") String concurrency,
+    /**
+     * Maximum recursion depth for archive extraction (default: 3).
+     * Set to 0 to disable recursive extraction (legacy behavior).
+     */
+    @Nullable @JsonProperty("max_archive_depth") Long maxArchiveDepth,
+    /**
+     * Tree-sitter language pack configuration (null = tree-sitter disabled).
+     *
+     * When set, enables code file extraction using tree-sitter parsers.
+     * Controls grammar download behavior and code analysis options.
+     */
+    @Nullable @JsonProperty("tree_sitter") TreeSitterConfig treeSitter,
+    /**
+     * Structured extraction via LLM (null = disabled).
+     *
+     * When set, the extracted document content is sent to an LLM with the
+     * provided JSON schema. The structured response is stored in
+     * {@code ExtractionResult.structured_output}.
+     */
+    @Nullable @JsonProperty("structured_extraction") StructuredExtractionConfig structuredExtraction,
+    /**
+     * Cancellation token for this extraction (null = no external cancellation).
+     *
+     * Pass a CancellationToken clone here and call CancellationToken.cancel
+     * from another thread / task to abort the extraction in progress. The extractor
+     * checks the token at safe checkpoints (before lock acquisition, between pages,
+     * between batch items) and returns KreuzbergError.Cancelled when set.
+     *
+     * The field is excluded from serialization because {@code CancellationToken} is a
+     * runtime handle, not a configuration value.
+     */
+    @Nullable @JsonProperty("cancel_token") String cancelToken
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("use_cache")
+private Boolean useCache = null;
+        @JsonProperty("enable_quality_processing")
+private Boolean enableQualityProcessing = null;
+        @Nullable private OcrConfig ocr = null;
+        @JsonProperty("force_ocr")
+private Boolean forceOcr = null;
+        @JsonProperty("force_ocr_pages")
+private List<Integer> forceOcrPages = null;
+        @JsonProperty("disable_ocr")
+private Boolean disableOcr = null;
+        @Nullable private ChunkingConfig chunking = null;
+        @JsonProperty("content_filter")
+        @Nullable private ContentFilterConfig contentFilter = null;
+        @Nullable private ImageExtractionConfig images = null;
+        @JsonProperty("pdf_options")
+        @Nullable private PdfConfig pdfOptions = null;
+        @JsonProperty("token_reduction")
+        @Nullable private TokenReductionOptions tokenReduction = null;
+        @JsonProperty("language_detection")
+        @Nullable private LanguageDetectionConfig languageDetection = null;
+        @Nullable private PageConfig pages = null;
+        @Nullable private KeywordConfig keywords = null;
+        @Nullable private PostProcessorConfig postprocessor = null;
+        @JsonProperty("html_options")
+private String htmlOptions = null;
+        @JsonProperty("html_output")
+        @Nullable private HtmlOutputConfig htmlOutput = null;
+        @JsonProperty("extraction_timeout_secs")
+private Long extractionTimeoutSecs = null;
+        @JsonProperty("max_concurrent_extractions")
+private Long maxConcurrentExtractions = null;
+        @JsonProperty("result_format")
+        @Nullable private ResultFormat resultFormat = ResultFormat.Unified;
+        @JsonProperty("security_limits")
+        @Nullable private SecurityLimits securityLimits = null;
+        @JsonProperty("max_embedded_file_bytes")
+private Long maxEmbeddedFileBytes = null;
+        @JsonProperty("output_format")
+        @Nullable private OutputFormat outputFormat = OutputFormat.Plain;
+        @Nullable private LayoutDetectionConfig layout = null;
+        @JsonProperty("use_layout_for_markdown")
+private Boolean useLayoutForMarkdown = null;
+        @JsonProperty("include_document_structure")
+private Boolean includeDocumentStructure = null;
+        @Nullable private AccelerationConfig acceleration = null;
+        @JsonProperty("cache_namespace")
+private String cacheNamespace = null;
+        @JsonProperty("cache_ttl_secs")
+private Long cacheTtlSecs = null;
+        @Nullable private EmailConfig email = null;
+private String concurrency = null;
+        @JsonProperty("max_archive_depth")
+private Long maxArchiveDepth = null;
+        @JsonProperty("tree_sitter")
+        @Nullable private TreeSitterConfig treeSitter = null;
+        @JsonProperty("structured_extraction")
+        @Nullable private StructuredExtractionConfig structuredExtraction = null;
+        @JsonProperty("cancel_token")
+private String cancelToken = null;
+
+        /** Sets the useCache field. */
+        @JsonProperty("use_cache")
+        public Builder withUseCache(final @Nullable Boolean value) {
+            this.useCache = value;
+            return this;
+        }
+
+        /** Sets the enableQualityProcessing field. */
+        @JsonProperty("enable_quality_processing")
+        public Builder withEnableQualityProcessing(final @Nullable Boolean value) {
+            this.enableQualityProcessing = value;
+            return this;
+        }
+
+        /** Sets the ocr field. */
+        @JsonProperty("ocr")
+        public Builder withOcr(final @Nullable OcrConfig value) {
+            this.ocr = value;
+            return this;
+        }
+
+        /** Sets the forceOcr field. */
+        @JsonProperty("force_ocr")
+        public Builder withForceOcr(final @Nullable Boolean value) {
+            this.forceOcr = value;
+            return this;
+        }
+
+        /** Sets the forceOcrPages field. */
+        @JsonProperty("force_ocr_pages")
+        public Builder withForceOcrPages(final @Nullable List<Integer> value) {
+            this.forceOcrPages = value;
+            return this;
+        }
+
+        /** Sets the disableOcr field. */
+        @JsonProperty("disable_ocr")
+        public Builder withDisableOcr(final @Nullable Boolean value) {
+            this.disableOcr = value;
+            return this;
+        }
+
+        /** Sets the chunking field. */
+        @JsonProperty("chunking")
+        public Builder withChunking(final @Nullable ChunkingConfig value) {
+            this.chunking = value;
+            return this;
+        }
+
+        /** Sets the contentFilter field. */
+        @JsonProperty("content_filter")
+        public Builder withContentFilter(final @Nullable ContentFilterConfig value) {
+            this.contentFilter = value;
+            return this;
+        }
+
+        /** Sets the images field. */
+        @JsonProperty("images")
+        public Builder withImages(final @Nullable ImageExtractionConfig value) {
+            this.images = value;
+            return this;
+        }
+
+        /** Sets the pdfOptions field. */
+        @JsonProperty("pdf_options")
+        public Builder withPdfOptions(final @Nullable PdfConfig value) {
+            this.pdfOptions = value;
+            return this;
+        }
+
+        /** Sets the tokenReduction field. */
+        @JsonProperty("token_reduction")
+        public Builder withTokenReduction(final @Nullable TokenReductionOptions value) {
+            this.tokenReduction = value;
+            return this;
+        }
+
+        /** Sets the languageDetection field. */
+        @JsonProperty("language_detection")
+        public Builder withLanguageDetection(final @Nullable LanguageDetectionConfig value) {
+            this.languageDetection = value;
+            return this;
+        }
+
+        /** Sets the pages field. */
+        @JsonProperty("pages")
+        public Builder withPages(final @Nullable PageConfig value) {
+            this.pages = value;
+            return this;
+        }
+
+        /** Sets the keywords field. */
+        @JsonProperty("keywords")
+        public Builder withKeywords(final @Nullable KeywordConfig value) {
+            this.keywords = value;
+            return this;
+        }
+
+        /** Sets the postprocessor field. */
+        @JsonProperty("postprocessor")
+        public Builder withPostprocessor(final @Nullable PostProcessorConfig value) {
+            this.postprocessor = value;
+            return this;
+        }
+
+        /** Sets the htmlOptions field. */
+        @JsonProperty("html_options")
+        public Builder withHtmlOptions(final @Nullable String value) {
+            this.htmlOptions = value;
+            return this;
+        }
+
+        /** Sets the htmlOutput field. */
+        @JsonProperty("html_output")
+        public Builder withHtmlOutput(final @Nullable HtmlOutputConfig value) {
+            this.htmlOutput = value;
+            return this;
+        }
+
+        /** Sets the extractionTimeoutSecs field. */
+        @JsonProperty("extraction_timeout_secs")
+        public Builder withExtractionTimeoutSecs(final @Nullable Long value) {
+            this.extractionTimeoutSecs = value;
+            return this;
+        }
+
+        /** Sets the maxConcurrentExtractions field. */
+        @JsonProperty("max_concurrent_extractions")
+        public Builder withMaxConcurrentExtractions(final @Nullable Long value) {
+            this.maxConcurrentExtractions = value;
+            return this;
+        }
+
+        /** Sets the resultFormat field. */
+        @JsonProperty("result_format")
+        public Builder withResultFormat(final @Nullable ResultFormat value) {
+            this.resultFormat = value;
+            return this;
+        }
+
+        /** Sets the securityLimits field. */
+        @JsonProperty("security_limits")
+        public Builder withSecurityLimits(final @Nullable SecurityLimits value) {
+            this.securityLimits = value;
+            return this;
+        }
+
+        /** Sets the maxEmbeddedFileBytes field. */
+        @JsonProperty("max_embedded_file_bytes")
+        public Builder withMaxEmbeddedFileBytes(final @Nullable Long value) {
+            this.maxEmbeddedFileBytes = value;
+            return this;
+        }
+
+        /** Sets the outputFormat field. */
+        @JsonProperty("output_format")
+        public Builder withOutputFormat(final @Nullable OutputFormat value) {
+            this.outputFormat = value;
+            return this;
+        }
+
+        /** Sets the layout field. */
+        @JsonProperty("layout")
+        public Builder withLayout(final @Nullable LayoutDetectionConfig value) {
+            this.layout = value;
+            return this;
+        }
+
+        /** Sets the useLayoutForMarkdown field. */
+        @JsonProperty("use_layout_for_markdown")
+        public Builder withUseLayoutForMarkdown(final @Nullable Boolean value) {
+            this.useLayoutForMarkdown = value;
+            return this;
+        }
+
+        /** Sets the includeDocumentStructure field. */
+        @JsonProperty("include_document_structure")
+        public Builder withIncludeDocumentStructure(final @Nullable Boolean value) {
+            this.includeDocumentStructure = value;
+            return this;
+        }
+
+        /** Sets the acceleration field. */
+        @JsonProperty("acceleration")
+        public Builder withAcceleration(final @Nullable AccelerationConfig value) {
+            this.acceleration = value;
+            return this;
+        }
+
+        /** Sets the cacheNamespace field. */
+        @JsonProperty("cache_namespace")
+        public Builder withCacheNamespace(final @Nullable String value) {
+            this.cacheNamespace = value;
+            return this;
+        }
+
+        /** Sets the cacheTtlSecs field. */
+        @JsonProperty("cache_ttl_secs")
+        public Builder withCacheTtlSecs(final @Nullable Long value) {
+            this.cacheTtlSecs = value;
+            return this;
+        }
+
+        /** Sets the email field. */
+        @JsonProperty("email")
+        public Builder withEmail(final @Nullable EmailConfig value) {
+            this.email = value;
+            return this;
+        }
+
+        /** Sets the concurrency field. */
+        @JsonProperty("concurrency")
+        public Builder withConcurrency(final @Nullable String value) {
+            this.concurrency = value;
+            return this;
+        }
+
+        /** Sets the maxArchiveDepth field. */
+        @JsonProperty("max_archive_depth")
+        public Builder withMaxArchiveDepth(final @Nullable Long value) {
+            this.maxArchiveDepth = value;
+            return this;
+        }
+
+        /** Sets the treeSitter field. */
+        @JsonProperty("tree_sitter")
+        public Builder withTreeSitter(final @Nullable TreeSitterConfig value) {
+            this.treeSitter = value;
+            return this;
+        }
+
+        /** Sets the structuredExtraction field. */
+        @JsonProperty("structured_extraction")
+        public Builder withStructuredExtraction(final @Nullable StructuredExtractionConfig value) {
+            this.structuredExtraction = value;
+            return this;
+        }
+
+        /** Sets the cancelToken field. */
+        @JsonProperty("cancel_token")
+        public Builder withCancelToken(final @Nullable String value) {
+            this.cancelToken = value;
+            return this;
+        }
+
+        /** Builds the ExtractionConfig instance. */
+        public ExtractionConfig build() {
+            return new ExtractionConfig(
+                useCache,
+                enableQualityProcessing,
+                ocr,
+                forceOcr,
+                forceOcrPages,
+                disableOcr,
+                chunking,
+                contentFilter,
+                images,
+                pdfOptions,
+                tokenReduction,
+                languageDetection,
+                pages,
+                keywords,
+                postprocessor,
+                htmlOptions,
+                htmlOutput,
+                extractionTimeoutSecs,
+                maxConcurrentExtractions,
+                resultFormat,
+                securityLimits,
+                maxEmbeddedFileBytes,
+                outputFormat,
+                layout,
+                useLayoutForMarkdown,
+                includeDocumentStructure,
+                acceleration,
+                cacheNamespace,
+                cacheTtlSecs,
+                email,
+                concurrency,
+                maxArchiveDepth,
+                treeSitter,
+                structuredExtraction,
+                cancelToken
+            );
+        }
+    }
+    // CPD-ON
+    public static ExtractionConfig defaultInstance() {
+        throw new UnsupportedOperationException("defaultInstance is not yet bridged via JNI; use the Builder instead.");
+    }
+}

packages/java/dev/kreuzberg/ExtractionDiff.java

@@ -0,0 +1,132 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.databind.JsonNode;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * The complete diff between two {@code ExtractionResult} values.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ExtractionDiff.Builder.class)
+public record ExtractionDiff(
+    /**
+     * Unified-diff hunks for the {@code content} field.
+     *
+     * Empty when the content is identical.
+     */
+    @JsonProperty("content_diff") List<DiffHunk> contentDiff,
+    /**
+     * Tables present in {@code b} but not in {@code a} (by index position, excess right-side tables).
+     */
+    @JsonProperty("tables_added") List<Table> tablesAdded,
+    /**
+     * Tables present in {@code a} but not in {@code b} (by index position, excess left-side tables).
+     */
+    @JsonProperty("tables_removed") List<Table> tablesRemoved,
+    /**
+     * Cell-level changes for table pairs that share the same index and dimensions.
+     */
+    @JsonProperty("tables_changed") List<TableDiff> tablesChanged,
+    /**
+     * Metadata difference, encoded as a JSON object with three top-level keys:
+     * {@code added} (keys present in {@code b} but not {@code a}), {@code removed} (keys present in {@code a}
+     * but not {@code b}), and {@code changed} (keys whose values differ — each entry is
+     * {@code { "from": &lt;value-in-a&gt;, "to": &lt;value-in-b&gt; }}).
+     *
+     * This is NOT RFC 6902 JSON Patch — we deliberately chose a flatter shape
+     * to avoid pulling in a json-patch crate. If you need RFC 6902 semantics
+     * (with JSON Pointer paths) feed {@code a.metadata} and {@code b.metadata} to your
+     * preferred json-patch impl directly.
+     */
+    @JsonProperty("metadata_changed") JsonNode metadataChanged,
+    /**
+     * Changes to embedded archive children.
+     */
+    @JsonProperty("embedded_changes") EmbeddedChanges embeddedChanges
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("content_diff")
+private List<DiffHunk> contentDiff = List.of();
+        @JsonProperty("tables_added")
+private List<Table> tablesAdded = List.of();
+        @JsonProperty("tables_removed")
+private List<Table> tablesRemoved = List.of();
+        @JsonProperty("tables_changed")
+private List<TableDiff> tablesChanged = List.of();
+        @JsonProperty("metadata_changed")
+private JsonNode metadataChanged = null;
+        @JsonProperty("embedded_changes")
+private EmbeddedChanges embeddedChanges = null;
+
+        /** Sets the contentDiff field. */
+        @JsonProperty("content_diff")
+        public Builder withContentDiff(final List<DiffHunk> value) {
+            this.contentDiff = value;
+            return this;
+        }
+
+        /** Sets the tablesAdded field. */
+        @JsonProperty("tables_added")
+        public Builder withTablesAdded(final List<Table> value) {
+            this.tablesAdded = value;
+            return this;
+        }
+
+        /** Sets the tablesRemoved field. */
+        @JsonProperty("tables_removed")
+        public Builder withTablesRemoved(final List<Table> value) {
+            this.tablesRemoved = value;
+            return this;
+        }
+
+        /** Sets the tablesChanged field. */
+        @JsonProperty("tables_changed")
+        public Builder withTablesChanged(final List<TableDiff> value) {
+            this.tablesChanged = value;
+            return this;
+        }
+
+        /** Sets the metadataChanged field. */
+        @JsonProperty("metadata_changed")
+        public Builder withMetadataChanged(final JsonNode value) {
+            this.metadataChanged = value;
+            return this;
+        }
+
+        /** Sets the embeddedChanges field. */
+        @JsonProperty("embedded_changes")
+        public Builder withEmbeddedChanges(final EmbeddedChanges value) {
+            this.embeddedChanges = value;
+            return this;
+        }
+
+        /** Builds the ExtractionDiff instance. */
+        public ExtractionDiff build() {
+            return new ExtractionDiff(
+                contentDiff,
+                tablesAdded,
+                tablesRemoved,
+                tablesChanged,
+                metadataChanged,
+                embeddedChanges
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/ExtractionMethod.java

@@ -0,0 +1,48 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonCreator;
+import com.fasterxml.jackson.annotation.JsonValue;
+
+/**
+ * How the extracted text was produced.
+ */
+public enum ExtractionMethod {
+    Native("native"),
+    Ocr("ocr"),
+    Mixed("mixed");
+
+    /** The string value. */
+    private final String value;
+
+    ExtractionMethod(final String value) {
+        this.value = value;
+    }
+
+    /** Returns the string value. */
+    @JsonValue
+    public String getValue() {
+        return value;
+    }
+
+    /** Creates an instance from a string value. */
+    @JsonCreator
+    public static ExtractionMethod fromValue(final String value) {
+        for (ExtractionMethod e : values()) {
+            if (e.value.equalsIgnoreCase(value)) {
+                return e;
+            }
+        }
+        throw new IllegalArgumentException("Unknown value: " + value);
+    }
+
+    /** Returns the wire-format string value (matches JSON serialization). */
+    @Override
+    public String toString() {
+        return value;
+    }
+}

packages/java/dev/kreuzberg/ExtractionResult.java

@@ -0,0 +1,480 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.databind.JsonNode;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * General extraction result used by the core extraction API.
+ *
+ * This is the main result type returned by all extraction functions.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ExtractionResult.Builder.class)
+public record ExtractionResult(
+    @JsonProperty("content") String content,
+    @JsonProperty("mime_type") String mimeType,
+    @JsonProperty("metadata") Metadata metadata,
+    /**
+     * Extraction strategy used to produce the returned text.
+     *
+     * Populated when the extractor can reliably distinguish native text extraction,
+     * OCR-only extraction, or mixed native/OCR output.
+     */
+    @Nullable @JsonProperty("extraction_method") ExtractionMethod extractionMethod,
+    @JsonProperty("tables") List<Table> tables,
+    @Nullable @JsonProperty("detected_languages") List<String> detectedLanguages,
+    /**
+     * Text chunks when chunking is enabled.
+     *
+     * When chunking configuration is provided, the content is split into
+     * overlapping chunks for efficient processing. Each chunk contains the text,
+     * optional embeddings (if enabled), and metadata about its position.
+     */
+    @Nullable @JsonProperty("chunks") List<Chunk> chunks,
+    /**
+     * Extracted images from the document.
+     *
+     * When image extraction is enabled via {@code ImageExtractionConfig}, this field
+     * contains all images found in the document with their raw data and metadata.
+     * Each image may optionally contain a nested {@code ocr_result} if OCR was performed.
+     */
+    @Nullable @JsonProperty("images") List<ExtractedImage> images,
+    /**
+     * Per-page content when page extraction is enabled.
+     *
+     * When page extraction is configured, the document is split into per-page content
+     * with tables and images mapped to their respective pages.
+     */
+    @Nullable @JsonProperty("pages") List<PageContent> pages,
+    /**
+     * Semantic elements when element-based result format is enabled.
+     *
+     * When result_format is set to ElementBased, this field contains semantic
+     * elements with type classification, unique identifiers, and metadata for
+     * Unstructured-compatible element-based processing.
+     */
+    @Nullable @JsonProperty("elements") List<Element> elements,
+    /**
+     * Rich Djot content structure (when extracting Djot documents).
+     *
+     * When extracting Djot documents with structured extraction enabled,
+     * this field contains the full semantic structure including:
+     * - Block-level elements with nesting
+     * - Inline formatting with attributes
+     * - Links, images, footnotes
+     * - Math expressions
+     * - Complete attribute information
+     *
+     * The {@code content} field still contains plain text for backward compatibility.
+     *
+     * Always {@code None} for non-Djot documents.
+     */
+    @Nullable @JsonProperty("djot_content") DjotContent djotContent,
+    /**
+     * OCR elements with full spatial and confidence metadata.
+     *
+     * When OCR is performed with element extraction enabled, this field contains
+     * the structured representation of detected text including:
+     * - Bounding geometry (rectangles or quadrilaterals)
+     * - Confidence scores (detection and recognition)
+     * - Rotation information
+     * - Hierarchical relationships (Tesseract only)
+     *
+     * This field preserves all metadata that would otherwise be lost when
+     * converting to plain text or markdown output formats.
+     *
+     * Only populated when {@code OcrElementConfig.include_elements} is true.
+     */
+    @Nullable @JsonProperty("ocr_elements") List<OcrElement> ocrElements,
+    /**
+     * Structured document tree (when document structure extraction is enabled).
+     *
+     * When {@code include_document_structure} is true in {@code ExtractionConfig}, this field
+     * contains the full hierarchical representation of the document including:
+     * - Heading-driven section nesting
+     * - Table grids with cell-level metadata
+     * - Content layer classification (body, header, footer, footnote)
+     * - Inline text annotations (formatting, links)
+     * - Bounding boxes and page numbers
+     *
+     * Independent of {@code result_format} — can be combined with Unified or ElementBased.
+     */
+    @Nullable @JsonProperty("document") DocumentStructure document,
+    /**
+     * Extracted keywords when keyword extraction is enabled.
+     *
+     * When keyword extraction (RAKE or YAKE) is configured, this field contains
+     * the extracted keywords with scores, algorithm info, and position data.
+     * Previously stored in {@code metadata.additional["keywords"]}.
+     */
+    @Nullable @JsonProperty("extracted_keywords") List<Keyword> extractedKeywords,
+    /**
+     * Document quality score from quality analysis.
+     *
+     * A value between 0.0 and 1.0 indicating the overall text quality.
+     * Previously stored in {@code metadata.additional["quality_score"]}.
+     */
+    @Nullable @JsonProperty("quality_score") Double qualityScore,
+    /**
+     * Non-fatal warnings collected during processing pipeline stages.
+     *
+     * Captures errors from optional pipeline features (embedding, chunking,
+     * language detection, output formatting) that don't prevent extraction
+     * but may indicate degraded results.
+     * Previously stored as individual keys in {@code metadata.additional}.
+     */
+    @Nullable @JsonProperty("processing_warnings") List<ProcessingWarning> processingWarnings,
+    /**
+     * PDF annotations extracted from the document.
+     *
+     * When annotation extraction is enabled via {@code PdfConfig.extract_annotations},
+     * this field contains text notes, highlights, links, stamps, and other
+     * annotations found in PDF documents.
+     */
+    @Nullable @JsonProperty("annotations") List<PdfAnnotation> annotations,
+    /**
+     * Nested extraction results from archive contents.
+     *
+     * When extracting archives, each processable file inside produces its own
+     * full extraction result. Set to {@code None} for non-archive formats.
+     * Use {@code max_archive_depth} in config to control recursion depth.
+     */
+    @Nullable @JsonProperty("children") List<ArchiveEntry> children,
+    /**
+     * URIs/links discovered during document extraction.
+     *
+     * Contains hyperlinks, image references, citations, email addresses, and
+     * other URI-like references found in the document. Always extracted when
+     * present in the source document.
+     */
+    @Nullable @JsonProperty("uris") List<ExtractedUri> uris,
+    /**
+     * Tracked changes embedded in the source document.
+     *
+     * Populated by per-format extractors that understand change-tracking
+     * metadata (DOCX {@code w:ins}/{@code w:del}/{@code w:rPrChange}, ODT {@code text:change-*},
+     * …). Every extractor defaults to {@code None} until its format-specific
+     * implementation is added. Extractors that do populate this field follow
+     * the "accepted-changes" convention: inserted text is present in
+     * {@code content}, deleted text is absent — the revision list is the separate
+     * audit trail.
+     */
+    @Nullable @JsonProperty("revisions") List<DocumentRevision> revisions,
+    /**
+     * Structured extraction output from LLM-based JSON schema extraction.
+     *
+     * When {@code structured_extraction} is configured in {@code ExtractionConfig}, the
+     * extracted document content is sent to a VLM with the provided JSON schema.
+     * The response is parsed and stored here as a JSON value matching the schema.
+     */
+    @Nullable @JsonProperty("structured_output") JsonNode structuredOutput,
+    /**
+     * Code intelligence results from tree-sitter analysis.
+     *
+     * Populated when extracting source code files with the {@code tree-sitter} feature.
+     * Contains metrics, structural analysis, imports/exports, comments,
+     * docstrings, symbols, diagnostics, and optionally chunked code segments.
+     *
+     * Stored as an opaque JSON value so that all language bindings (Go, Java,
+     * C#, …) can deserialize it as a raw JSON object rather than a typed struct.
+     * The underlying type is {@code tree_sitter_language_pack.ProcessResult}.
+     */
+    @Nullable @JsonProperty("code_intelligence") JsonNode codeIntelligence,
+    /**
+     * LLM token usage and cost data for all LLM calls made during this extraction.
+     *
+     * Contains one entry per LLM call. Multiple entries are produced when
+     * VLM OCR, structured extraction, or LLM embeddings run during
+     * the same extraction.
+     *
+     * {@code None} when no LLM was used.
+     */
+    @Nullable @JsonProperty("llm_usage") List<LlmUsage> llmUsage,
+    /**
+     * Pre-rendered content in the requested output format.
+     *
+     * Populated during {@code derive_extraction_result} before tree derivation consumes
+     * element data. {@code apply_output_format} swaps this into {@code content} at the end
+     * of the pipeline, after post-processors have operated on plain text.
+     */
+    @Nullable @JsonProperty("formatted_content") String formattedContent,
+    /**
+     * Structured hOCR document for the OCR+layout pipeline.
+     *
+     * When tesseract produces hOCR output, the parsed {@code InternalDocument} carries
+     * paragraph structure with bounding boxes and confidence scores. The layout
+     * classification step enriches these elements before final rendering.
+     */
+    @Nullable @JsonProperty("ocr_internal_document") String ocrInternalDocument
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String content = "";
+        @JsonProperty("mime_type")
+private String mimeType = "";
+private Metadata metadata = null;
+        @JsonProperty("extraction_method")
+        @Nullable private ExtractionMethod extractionMethod = null;
+private List<Table> tables = List.of();
+        @JsonProperty("detected_languages")
+private List<String> detectedLanguages = null;
+private List<Chunk> chunks = null;
+private List<ExtractedImage> images = null;
+private List<PageContent> pages = null;
+private List<Element> elements = null;
+        @JsonProperty("djot_content")
+        @Nullable private DjotContent djotContent = null;
+        @JsonProperty("ocr_elements")
+private List<OcrElement> ocrElements = null;
+        @Nullable private DocumentStructure document = null;
+        @JsonProperty("extracted_keywords")
+private List<Keyword> extractedKeywords = null;
+        @JsonProperty("quality_score")
+private Double qualityScore = null;
+        @JsonProperty("processing_warnings")
+private List<ProcessingWarning> processingWarnings = null;
+private List<PdfAnnotation> annotations = null;
+private List<ArchiveEntry> children = null;
+private List<ExtractedUri> uris = null;
+private List<DocumentRevision> revisions = null;
+        @JsonProperty("structured_output")
+private JsonNode structuredOutput = null;
+        @JsonProperty("code_intelligence")
+private JsonNode codeIntelligence = null;
+        @JsonProperty("llm_usage")
+private List<LlmUsage> llmUsage = null;
+        @JsonProperty("formatted_content")
+private String formattedContent = null;
+        @JsonProperty("ocr_internal_document")
+private String ocrInternalDocument = null;
+
+        /** Sets the content field. */
+        @JsonProperty("content")
+        public Builder withContent(final String value) {
+            this.content = value;
+            return this;
+        }
+
+        /** Sets the mimeType field. */
+        @JsonProperty("mime_type")
+        public Builder withMimeType(final String value) {
+            this.mimeType = value;
+            return this;
+        }
+
+        /** Sets the metadata field. */
+        @JsonProperty("metadata")
+        public Builder withMetadata(final Metadata value) {
+            this.metadata = value;
+            return this;
+        }
+
+        /** Sets the extractionMethod field. */
+        @JsonProperty("extraction_method")
+        public Builder withExtractionMethod(final @Nullable ExtractionMethod value) {
+            this.extractionMethod = value;
+            return this;
+        }
+
+        /** Sets the tables field. */
+        @JsonProperty("tables")
+        public Builder withTables(final List<Table> value) {
+            this.tables = value;
+            return this;
+        }
+
+        /** Sets the detectedLanguages field. */
+        @JsonProperty("detected_languages")
+        public Builder withDetectedLanguages(final @Nullable List<String> value) {
+            this.detectedLanguages = value;
+            return this;
+        }
+
+        /** Sets the chunks field. */
+        @JsonProperty("chunks")
+        public Builder withChunks(final @Nullable List<Chunk> value) {
+            this.chunks = value;
+            return this;
+        }
+
+        /** Sets the images field. */
+        @JsonProperty("images")
+        public Builder withImages(final @Nullable List<ExtractedImage> value) {
+            this.images = value;
+            return this;
+        }
+
+        /** Sets the pages field. */
+        @JsonProperty("pages")
+        public Builder withPages(final @Nullable List<PageContent> value) {
+            this.pages = value;
+            return this;
+        }
+
+        /** Sets the elements field. */
+        @JsonProperty("elements")
+        public Builder withElements(final @Nullable List<Element> value) {
+            this.elements = value;
+            return this;
+        }
+
+        /** Sets the djotContent field. */
+        @JsonProperty("djot_content")
+        public Builder withDjotContent(final @Nullable DjotContent value) {
+            this.djotContent = value;
+            return this;
+        }
+
+        /** Sets the ocrElements field. */
+        @JsonProperty("ocr_elements")
+        public Builder withOcrElements(final @Nullable List<OcrElement> value) {
+            this.ocrElements = value;
+            return this;
+        }
+
+        /** Sets the document field. */
+        @JsonProperty("document")
+        public Builder withDocument(final @Nullable DocumentStructure value) {
+            this.document = value;
+            return this;
+        }
+
+        /** Sets the extractedKeywords field. */
+        @JsonProperty("extracted_keywords")
+        public Builder withExtractedKeywords(final @Nullable List<Keyword> value) {
+            this.extractedKeywords = value;
+            return this;
+        }
+
+        /** Sets the qualityScore field. */
+        @JsonProperty("quality_score")
+        public Builder withQualityScore(final @Nullable Double value) {
+            this.qualityScore = value;
+            return this;
+        }
+
+        /** Sets the processingWarnings field. */
+        @JsonProperty("processing_warnings")
+        public Builder withProcessingWarnings(final @Nullable List<ProcessingWarning> value) {
+            this.processingWarnings = value;
+            return this;
+        }
+
+        /** Sets the annotations field. */
+        @JsonProperty("annotations")
+        public Builder withAnnotations(final @Nullable List<PdfAnnotation> value) {
+            this.annotations = value;
+            return this;
+        }
+
+        /** Sets the children field. */
+        @JsonProperty("children")
+        public Builder withChildren(final @Nullable List<ArchiveEntry> value) {
+            this.children = value;
+            return this;
+        }
+
+        /** Sets the uris field. */
+        @JsonProperty("uris")
+        public Builder withUris(final @Nullable List<ExtractedUri> value) {
+            this.uris = value;
+            return this;
+        }
+
+        /** Sets the revisions field. */
+        @JsonProperty("revisions")
+        public Builder withRevisions(final @Nullable List<DocumentRevision> value) {
+            this.revisions = value;
+            return this;
+        }
+
+        /** Sets the structuredOutput field. */
+        @JsonProperty("structured_output")
+        public Builder withStructuredOutput(final @Nullable JsonNode value) {
+            this.structuredOutput = value;
+            return this;
+        }
+
+        /** Sets the codeIntelligence field. */
+        @JsonProperty("code_intelligence")
+        public Builder withCodeIntelligence(final @Nullable JsonNode value) {
+            this.codeIntelligence = value;
+            return this;
+        }
+
+        /** Sets the llmUsage field. */
+        @JsonProperty("llm_usage")
+        public Builder withLlmUsage(final @Nullable List<LlmUsage> value) {
+            this.llmUsage = value;
+            return this;
+        }
+
+        /** Sets the formattedContent field. */
+        @JsonProperty("formatted_content")
+        public Builder withFormattedContent(final @Nullable String value) {
+            this.formattedContent = value;
+            return this;
+        }
+
+        /** Sets the ocrInternalDocument field. */
+        @JsonProperty("ocr_internal_document")
+        public Builder withOcrInternalDocument(final @Nullable String value) {
+            this.ocrInternalDocument = value;
+            return this;
+        }
+
+        /** Builds the ExtractionResult instance. */
+        public ExtractionResult build() {
+            return new ExtractionResult(
+                content,
+                mimeType,
+                metadata,
+                extractionMethod,
+                tables,
+                detectedLanguages,
+                chunks,
+                images,
+                pages,
+                elements,
+                djotContent,
+                ocrElements,
+                document,
+                extractedKeywords,
+                qualityScore,
+                processingWarnings,
+                annotations,
+                children,
+                uris,
+                revisions,
+                structuredOutput,
+                codeIntelligence,
+                llmUsage,
+                formattedContent,
+                ocrInternalDocument
+            );
+        }
+    }
+    // CPD-ON
+    /**
+     * Convert from an OCR result.
+     */
+    public static ExtractionResult fromOcr(OcrExtractionResult ocr) {
+        throw new UnsupportedOperationException("fromOcr is not yet bridged via JNI; use the Builder instead.");
+    }
+}

packages/java/dev/kreuzberg/FictionBookMetadata.java

@@ -0,0 +1,68 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * FictionBook (FB2) metadata.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = FictionBookMetadata.Builder.class)
+public record FictionBookMetadata(
+    @Nullable @JsonProperty("genres") List<String> genres,
+    @Nullable @JsonProperty("sequences") List<String> sequences,
+    @Nullable @JsonProperty("annotation") String annotation
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private List<String> genres = null;
+private List<String> sequences = null;
+private String annotation = null;
+
+        /** Sets the genres field. */
+        @JsonProperty("genres")
+        public Builder withGenres(final @Nullable List<String> value) {
+            this.genres = value;
+            return this;
+        }
+
+        /** Sets the sequences field. */
+        @JsonProperty("sequences")
+        public Builder withSequences(final @Nullable List<String> value) {
+            this.sequences = value;
+            return this;
+        }
+
+        /** Sets the annotation field. */
+        @JsonProperty("annotation")
+        public Builder withAnnotation(final @Nullable String value) {
+            this.annotation = value;
+            return this;
+        }
+
+        /** Builds the FictionBookMetadata instance. */
+        public FictionBookMetadata build() {
+            return new FictionBookMetadata(
+                genres,
+                sequences,
+                annotation
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/FileExtractionConfig.java

@@ -0,0 +1,361 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Per-file extraction configuration overrides for batch processing.
+ *
+ * All fields are {@code Option<T>} — {@code None} means "use the batch-level default."
+ * This type is used with {@code batch_extract_files} and
+ * {@code batch_extract_bytes} to allow heterogeneous
+ * extraction settings within a single batch.
+ *
+ * # Excluded Fields
+ *
+ * The following {@code ExtractionConfig} fields are batch-level only and
+ * cannot be overridden per file:
+ * - {@code max_concurrent_extractions} — controls batch parallelism
+ * - {@code use_cache} — global caching policy
+ * - {@code acceleration} — shared ONNX execution provider
+ * - {@code security_limits} — global archive security policy
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = FileExtractionConfig.Builder.class)
+public record FileExtractionConfig(
+    /**
+     * Override quality post-processing for this file.
+     */
+    @Nullable @JsonProperty("enable_quality_processing") Boolean enableQualityProcessing,
+    /**
+     * Override OCR configuration for this file (null in the Option = use batch default).
+     */
+    @Nullable @JsonProperty("ocr") OcrConfig ocr,
+    /**
+     * Override force OCR for this file.
+     */
+    @Nullable @JsonProperty("force_ocr") Boolean forceOcr,
+    /**
+     * Override force OCR pages for this file (1-indexed page numbers).
+     */
+    @Nullable @JsonProperty("force_ocr_pages") List<Integer> forceOcrPages,
+    /**
+     * Override disable OCR for this file.
+     */
+    @Nullable @JsonProperty("disable_ocr") Boolean disableOcr,
+    /**
+     * Override chunking configuration for this file.
+     */
+    @Nullable @JsonProperty("chunking") ChunkingConfig chunking,
+    /**
+     * Override content filtering configuration for this file.
+     */
+    @Nullable @JsonProperty("content_filter") ContentFilterConfig contentFilter,
+    /**
+     * Override image extraction configuration for this file.
+     */
+    @Nullable @JsonProperty("images") ImageExtractionConfig images,
+    /**
+     * Override PDF options for this file.
+     */
+    @Nullable @JsonProperty("pdf_options") PdfConfig pdfOptions,
+    /**
+     * Override token reduction for this file.
+     */
+    @Nullable @JsonProperty("token_reduction") TokenReductionOptions tokenReduction,
+    /**
+     * Override language detection for this file.
+     */
+    @Nullable @JsonProperty("language_detection") LanguageDetectionConfig languageDetection,
+    /**
+     * Override page extraction for this file.
+     */
+    @Nullable @JsonProperty("pages") PageConfig pages,
+    /**
+     * Override keyword extraction for this file.
+     */
+    @Nullable @JsonProperty("keywords") KeywordConfig keywords,
+    /**
+     * Override post-processor for this file.
+     */
+    @Nullable @JsonProperty("postprocessor") PostProcessorConfig postprocessor,
+    /**
+     * Override HTML conversion options for this file.
+     */
+    @Nullable @JsonProperty("html_options") String htmlOptions,
+    /**
+     * Override result format for this file.
+     */
+    @Nullable @JsonProperty("result_format") ResultFormat resultFormat,
+    /**
+     * Override output content format for this file.
+     */
+    @Nullable @JsonProperty("output_format") OutputFormat outputFormat,
+    /**
+     * Override document structure output for this file.
+     */
+    @Nullable @JsonProperty("include_document_structure") Boolean includeDocumentStructure,
+    /**
+     * Override layout detection for this file.
+     */
+    @Nullable @JsonProperty("layout") LayoutDetectionConfig layout,
+    /**
+     * Override per-file extraction timeout in seconds.
+     *
+     * When set, the extraction for this file will be canceled after the
+     * specified duration. A timed-out file produces an error result without
+     * affecting other files in the batch.
+     */
+    @Nullable @JsonProperty("timeout_secs") Long timeoutSecs,
+    /**
+     * Override tree-sitter configuration for this file.
+     */
+    @Nullable @JsonProperty("tree_sitter") TreeSitterConfig treeSitter,
+    /**
+     * Override structured extraction configuration for this file.
+     *
+     * When set, enables LLM-based structured extraction with a JSON schema
+     * for this specific file. The extracted content is sent to a VLM/LLM
+     * and the response is parsed according to the provided schema.
+     */
+    @Nullable @JsonProperty("structured_extraction") StructuredExtractionConfig structuredExtraction
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("enable_quality_processing")
+private Boolean enableQualityProcessing = null;
+private OcrConfig ocr = null;
+        @JsonProperty("force_ocr")
+private Boolean forceOcr = null;
+        @JsonProperty("force_ocr_pages")
+private List<Integer> forceOcrPages = null;
+        @JsonProperty("disable_ocr")
+private Boolean disableOcr = null;
+private ChunkingConfig chunking = null;
+        @JsonProperty("content_filter")
+private ContentFilterConfig contentFilter = null;
+private ImageExtractionConfig images = null;
+        @JsonProperty("pdf_options")
+private PdfConfig pdfOptions = null;
+        @JsonProperty("token_reduction")
+private TokenReductionOptions tokenReduction = null;
+        @JsonProperty("language_detection")
+private LanguageDetectionConfig languageDetection = null;
+private PageConfig pages = null;
+private KeywordConfig keywords = null;
+private PostProcessorConfig postprocessor = null;
+        @JsonProperty("html_options")
+private String htmlOptions = null;
+        @JsonProperty("result_format")
+private ResultFormat resultFormat = null;
+        @JsonProperty("output_format")
+private OutputFormat outputFormat = null;
+        @JsonProperty("include_document_structure")
+private Boolean includeDocumentStructure = null;
+private LayoutDetectionConfig layout = null;
+        @JsonProperty("timeout_secs")
+private Long timeoutSecs = null;
+        @JsonProperty("tree_sitter")
+private TreeSitterConfig treeSitter = null;
+        @JsonProperty("structured_extraction")
+private StructuredExtractionConfig structuredExtraction = null;
+
+        /** Sets the enableQualityProcessing field. */
+        @JsonProperty("enable_quality_processing")
+        public Builder withEnableQualityProcessing(final @Nullable boolean value) {
+            this.enableQualityProcessing = value;
+            return this;
+        }
+
+        /** Sets the ocr field. */
+        @JsonProperty("ocr")
+        public Builder withOcr(final @Nullable OcrConfig value) {
+            this.ocr = value;
+            return this;
+        }
+
+        /** Sets the forceOcr field. */
+        @JsonProperty("force_ocr")
+        public Builder withForceOcr(final @Nullable boolean value) {
+            this.forceOcr = value;
+            return this;
+        }
+
+        /** Sets the forceOcrPages field. */
+        @JsonProperty("force_ocr_pages")
+        public Builder withForceOcrPages(final @Nullable List<Integer> value) {
+            this.forceOcrPages = value;
+            return this;
+        }
+
+        /** Sets the disableOcr field. */
+        @JsonProperty("disable_ocr")
+        public Builder withDisableOcr(final @Nullable boolean value) {
+            this.disableOcr = value;
+            return this;
+        }
+
+        /** Sets the chunking field. */
+        @JsonProperty("chunking")
+        public Builder withChunking(final @Nullable ChunkingConfig value) {
+            this.chunking = value;
+            return this;
+        }
+
+        /** Sets the contentFilter field. */
+        @JsonProperty("content_filter")
+        public Builder withContentFilter(final @Nullable ContentFilterConfig value) {
+            this.contentFilter = value;
+            return this;
+        }
+
+        /** Sets the images field. */
+        @JsonProperty("images")
+        public Builder withImages(final @Nullable ImageExtractionConfig value) {
+            this.images = value;
+            return this;
+        }
+
+        /** Sets the pdfOptions field. */
+        @JsonProperty("pdf_options")
+        public Builder withPdfOptions(final @Nullable PdfConfig value) {
+            this.pdfOptions = value;
+            return this;
+        }
+
+        /** Sets the tokenReduction field. */
+        @JsonProperty("token_reduction")
+        public Builder withTokenReduction(final @Nullable TokenReductionOptions value) {
+            this.tokenReduction = value;
+            return this;
+        }
+
+        /** Sets the languageDetection field. */
+        @JsonProperty("language_detection")
+        public Builder withLanguageDetection(final @Nullable LanguageDetectionConfig value) {
+            this.languageDetection = value;
+            return this;
+        }
+
+        /** Sets the pages field. */
+        @JsonProperty("pages")
+        public Builder withPages(final @Nullable PageConfig value) {
+            this.pages = value;
+            return this;
+        }
+
+        /** Sets the keywords field. */
+        @JsonProperty("keywords")
+        public Builder withKeywords(final @Nullable KeywordConfig value) {
+            this.keywords = value;
+            return this;
+        }
+
+        /** Sets the postprocessor field. */
+        @JsonProperty("postprocessor")
+        public Builder withPostprocessor(final @Nullable PostProcessorConfig value) {
+            this.postprocessor = value;
+            return this;
+        }
+
+        /** Sets the htmlOptions field. */
+        @JsonProperty("html_options")
+        public Builder withHtmlOptions(final @Nullable String value) {
+            this.htmlOptions = value;
+            return this;
+        }
+
+        /** Sets the resultFormat field. */
+        @JsonProperty("result_format")
+        public Builder withResultFormat(final @Nullable ResultFormat value) {
+            this.resultFormat = value;
+            return this;
+        }
+
+        /** Sets the outputFormat field. */
+        @JsonProperty("output_format")
+        public Builder withOutputFormat(final @Nullable OutputFormat value) {
+            this.outputFormat = value;
+            return this;
+        }
+
+        /** Sets the includeDocumentStructure field. */
+        @JsonProperty("include_document_structure")
+        public Builder withIncludeDocumentStructure(final @Nullable boolean value) {
+            this.includeDocumentStructure = value;
+            return this;
+        }
+
+        /** Sets the layout field. */
+        @JsonProperty("layout")
+        public Builder withLayout(final @Nullable LayoutDetectionConfig value) {
+            this.layout = value;
+            return this;
+        }
+
+        /** Sets the timeoutSecs field. */
+        @JsonProperty("timeout_secs")
+        public Builder withTimeoutSecs(final @Nullable long value) {
+            this.timeoutSecs = value;
+            return this;
+        }
+
+        /** Sets the treeSitter field. */
+        @JsonProperty("tree_sitter")
+        public Builder withTreeSitter(final @Nullable TreeSitterConfig value) {
+            this.treeSitter = value;
+            return this;
+        }
+
+        /** Sets the structuredExtraction field. */
+        @JsonProperty("structured_extraction")
+        public Builder withStructuredExtraction(final @Nullable StructuredExtractionConfig value) {
+            this.structuredExtraction = value;
+            return this;
+        }
+
+        /** Builds the FileExtractionConfig instance. */
+        public FileExtractionConfig build() {
+            return new FileExtractionConfig(
+                enableQualityProcessing,
+                ocr,
+                forceOcr,
+                forceOcrPages,
+                disableOcr,
+                chunking,
+                contentFilter,
+                images,
+                pdfOptions,
+                tokenReduction,
+                languageDetection,
+                pages,
+                keywords,
+                postprocessor,
+                htmlOptions,
+                resultFormat,
+                outputFormat,
+                includeDocumentStructure,
+                layout,
+                timeoutSecs,
+                treeSitter,
+                structuredExtraction
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/Footnote.java

@@ -0,0 +1,63 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * Footnote in Djot.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = Footnote.Builder.class)
+public record Footnote(
+    /**
+     * Footnote label
+     */
+    @JsonProperty("label") String label,
+    /**
+     * Footnote content blocks
+     */
+    @JsonProperty("content") List<FormattedBlock> content
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String label = "";
+private List<FormattedBlock> content = List.of();
+
+        /** Sets the label field. */
+        @JsonProperty("label")
+        public Builder withLabel(final String value) {
+            this.label = value;
+            return this;
+        }
+
+        /** Sets the content field. */
+        @JsonProperty("content")
+        public Builder withContent(final List<FormattedBlock> value) {
+            this.content = value;
+            return this;
+        }
+
+        /** Builds the Footnote instance. */
+        public Footnote build() {
+            return new Footnote(
+                label,
+                content
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/FormatMetadata.java

@@ -0,0 +1,275 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.databind.deser.std.StdDeserializer;
+import com.fasterxml.jackson.databind.ser.std.StdSerializer;
+import com.fasterxml.jackson.core.JsonParser;
+import com.fasterxml.jackson.core.JsonGenerator;
+import com.fasterxml.jackson.databind.DeserializationContext;
+import com.fasterxml.jackson.databind.SerializerProvider;
+import com.fasterxml.jackson.databind.node.ObjectNode;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonSerialize;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Format-specific metadata (discriminated union).
+ *
+ * Only one format type can exist per extraction result. This provides
+ * type-safe, clean metadata without nested optionals.
+ */
+@com.fasterxml.jackson.annotation.JsonIgnoreProperties(ignoreUnknown = true)
+@JsonDeserialize(using = FormatMetadataDeserializer.class)
+@JsonSerialize(using = FormatMetadataSerializer.class)
+public sealed interface FormatMetadata {
+
+    record Pdf(PdfMetadata value) implements FormatMetadata { }
+
+    record Docx(DocxMetadata value) implements FormatMetadata { }
+
+    record Excel(ExcelMetadata value) implements FormatMetadata { }
+
+    record Email(EmailMetadata value) implements FormatMetadata { }
+
+    record Pptx(PptxMetadata value) implements FormatMetadata { }
+
+    record Archive(ArchiveMetadata value) implements FormatMetadata { }
+
+    record Image(ImageMetadata value) implements FormatMetadata { }
+
+    record Xml(XmlMetadata value) implements FormatMetadata { }
+
+    record Text(TextMetadata value) implements FormatMetadata { }
+
+    record Html(HtmlMetadata value) implements FormatMetadata { }
+
+    record Ocr(OcrMetadata value) implements FormatMetadata { }
+
+    record Csv(CsvMetadata value) implements FormatMetadata { }
+
+    record Bibtex(BibtexMetadata value) implements FormatMetadata { }
+
+    record Citation(CitationMetadata value) implements FormatMetadata { }
+
+    record FictionBook(FictionBookMetadata value) implements FormatMetadata { }
+
+    record Dbf(DbfMetadata value) implements FormatMetadata { }
+
+    record Jats(JatsMetadata value) implements FormatMetadata { }
+
+    record Epub(EpubMetadata value) implements FormatMetadata { }
+
+    record Pst(PstMetadata value) implements FormatMetadata { }
+
+    record Code(String value) implements FormatMetadata { }
+
+    /** Returns the Pdf data if this is a Pdf variant, otherwise null. */
+    default @Nullable PdfMetadata pdf() {
+        return this instanceof Pdf e ? e.value() : null;
+    }
+
+    /** Returns the Docx data if this is a Docx variant, otherwise null. */
+    default @Nullable DocxMetadata docx() {
+        return this instanceof Docx e ? e.value() : null;
+    }
+
+    /** Returns the Excel data if this is a Excel variant, otherwise null. */
+    default @Nullable ExcelMetadata excel() {
+        return this instanceof Excel e ? e.value() : null;
+    }
+
+    /** Returns the Email data if this is a Email variant, otherwise null. */
+    default @Nullable EmailMetadata email() {
+        return this instanceof Email e ? e.value() : null;
+    }
+
+    /** Returns the Pptx data if this is a Pptx variant, otherwise null. */
+    default @Nullable PptxMetadata pptx() {
+        return this instanceof Pptx e ? e.value() : null;
+    }
+
+    /** Returns the Archive data if this is a Archive variant, otherwise null. */
+    default @Nullable ArchiveMetadata archive() {
+        return this instanceof Archive e ? e.value() : null;
+    }
+
+    /** Returns the Image data if this is a Image variant, otherwise null. */
+    default @Nullable ImageMetadata image() {
+        return this instanceof Image e ? e.value() : null;
+    }
+
+    /** Returns the Xml data if this is a Xml variant, otherwise null. */
+    default @Nullable XmlMetadata xml() {
+        return this instanceof Xml e ? e.value() : null;
+    }
+
+    /** Returns the Text data if this is a Text variant, otherwise null. */
+    default @Nullable TextMetadata text() {
+        return this instanceof Text e ? e.value() : null;
+    }
+
+    /** Returns the Html data if this is a Html variant, otherwise null. */
+    default @Nullable HtmlMetadata html() {
+        return this instanceof Html e ? e.value() : null;
+    }
+
+    /** Returns the Ocr data if this is a Ocr variant, otherwise null. */
+    default @Nullable OcrMetadata ocr() {
+        return this instanceof Ocr e ? e.value() : null;
+    }
+
+    /** Returns the Csv data if this is a Csv variant, otherwise null. */
+    default @Nullable CsvMetadata csv() {
+        return this instanceof Csv e ? e.value() : null;
+    }
+
+    /** Returns the Bibtex data if this is a Bibtex variant, otherwise null. */
+    default @Nullable BibtexMetadata bibtex() {
+        return this instanceof Bibtex e ? e.value() : null;
+    }
+
+    /** Returns the Citation data if this is a Citation variant, otherwise null. */
+    default @Nullable CitationMetadata citation() {
+        return this instanceof Citation e ? e.value() : null;
+    }
+
+    /** Returns the FictionBook data if this is a FictionBook variant, otherwise null. */
+    default @Nullable FictionBookMetadata fictionBook() {
+        return this instanceof FictionBook e ? e.value() : null;
+    }
+
+    /** Returns the Dbf data if this is a Dbf variant, otherwise null. */
+    default @Nullable DbfMetadata dbf() {
+        return this instanceof Dbf e ? e.value() : null;
+    }
+
+    /** Returns the Jats data if this is a Jats variant, otherwise null. */
+    default @Nullable JatsMetadata jats() {
+        return this instanceof Jats e ? e.value() : null;
+    }
+
+    /** Returns the Epub data if this is a Epub variant, otherwise null. */
+    default @Nullable EpubMetadata epub() {
+        return this instanceof Epub e ? e.value() : null;
+    }
+
+    /** Returns the Pst data if this is a Pst variant, otherwise null. */
+    default @Nullable PstMetadata pst() {
+        return this instanceof Pst e ? e.value() : null;
+    }
+
+    /** Returns the Code data if this is a Code variant, otherwise null. */
+    default @Nullable String code() {
+        return this instanceof Code e ? e.value() : null;
+    }
+
+}
+
+// Custom deserializer for sealed interface with unwrapped variants
+class FormatMetadataDeserializer extends StdDeserializer<FormatMetadata> {
+    FormatMetadataDeserializer() {
+        super(FormatMetadata.class);
+    }
+
+    @Override
+    public FormatMetadata deserialize(JsonParser parser, DeserializationContext ctx)
+            throws java.io.IOException {
+        ObjectNode node = parser.getCodec().readTree(parser);
+        com.fasterxml.jackson.databind.JsonNode tagNode = node.get("format_type");
+        if (tagNode == null || tagNode.isNull()) {
+            throw new com.fasterxml.jackson.databind.JsonMappingException(
+                parser, "Missing discriminator field: format_type");
+        }
+        String tagValue = tagNode.asText();
+        node.remove("format_type");
+
+        return switch (tagValue) {
+            case "pdf" -> new FormatMetadata.Pdf(ctx.readTreeAsValue(node, PdfMetadata.class));
+            case "docx" -> new FormatMetadata.Docx(ctx.readTreeAsValue(node, DocxMetadata.class));
+            case "excel" -> new FormatMetadata.Excel(ctx.readTreeAsValue(node, ExcelMetadata.class));
+            case "email" -> new FormatMetadata.Email(ctx.readTreeAsValue(node, EmailMetadata.class));
+            case "pptx" -> new FormatMetadata.Pptx(ctx.readTreeAsValue(node, PptxMetadata.class));
+            case "archive" -> new FormatMetadata.Archive(ctx.readTreeAsValue(node, ArchiveMetadata.class));
+            case "image" -> new FormatMetadata.Image(ctx.readTreeAsValue(node, ImageMetadata.class));
+            case "xml" -> new FormatMetadata.Xml(ctx.readTreeAsValue(node, XmlMetadata.class));
+            case "text" -> new FormatMetadata.Text(ctx.readTreeAsValue(node, TextMetadata.class));
+            case "html" -> new FormatMetadata.Html(ctx.readTreeAsValue(node, HtmlMetadata.class));
+            case "ocr" -> new FormatMetadata.Ocr(ctx.readTreeAsValue(node, OcrMetadata.class));
+            case "csv" -> new FormatMetadata.Csv(ctx.readTreeAsValue(node, CsvMetadata.class));
+            case "bibtex" -> new FormatMetadata.Bibtex(ctx.readTreeAsValue(node, BibtexMetadata.class));
+            case "citation" -> new FormatMetadata.Citation(ctx.readTreeAsValue(node, CitationMetadata.class));
+            case "fiction_book" -> new FormatMetadata.FictionBook(ctx.readTreeAsValue(node, FictionBookMetadata.class));
+            case "dbf" -> new FormatMetadata.Dbf(ctx.readTreeAsValue(node, DbfMetadata.class));
+            case "jats" -> new FormatMetadata.Jats(ctx.readTreeAsValue(node, JatsMetadata.class));
+            case "epub" -> new FormatMetadata.Epub(ctx.readTreeAsValue(node, EpubMetadata.class));
+            case "pst" -> new FormatMetadata.Pst(ctx.readTreeAsValue(node, PstMetadata.class));
+            case "code" -> new FormatMetadata.Code(node.toString());
+            default -> throw new com.fasterxml.jackson.databind.JsonMappingException(
+                parser, "Unknown FormatMetadata discriminator: " + tagValue);
+        };
+    }
+}
+
+// Custom serializer for sealed interface with unwrapped variants — emits
+// the discriminator tag alongside the inner record's fields (flat object).
+class FormatMetadataSerializer extends StdSerializer<FormatMetadata> {
+    private static final com.fasterxml.jackson.databind.ObjectMapper MAPPER =
+        new com.fasterxml.jackson.databind.ObjectMapper()
+            .registerModule(new com.fasterxml.jackson.datatype.jdk8.Jdk8Module())
+            .setPropertyNamingStrategy(com.fasterxml.jackson.databind.PropertyNamingStrategies.SNAKE_CASE)
+            .setSerializationInclusion(com.fasterxml.jackson.annotation.JsonInclude.Include.NON_NULL);
+
+    FormatMetadataSerializer() {
+        super(FormatMetadata.class);
+    }
+
+    @Override
+    public void serialize(FormatMetadata value, JsonGenerator gen, SerializerProvider provider)
+            throws java.io.IOException {
+        String tag;
+        Object inner;if (value instanceof FormatMetadata.Pdf v) {
+            tag = "pdf";            inner = v.value();        }else if (value instanceof FormatMetadata.Docx v) {
+            tag = "docx";            inner = v.value();        }else if (value instanceof FormatMetadata.Excel v) {
+            tag = "excel";            inner = v.value();        }else if (value instanceof FormatMetadata.Email v) {
+            tag = "email";            inner = v.value();        }else if (value instanceof FormatMetadata.Pptx v) {
+            tag = "pptx";            inner = v.value();        }else if (value instanceof FormatMetadata.Archive v) {
+            tag = "archive";            inner = v.value();        }else if (value instanceof FormatMetadata.Image v) {
+            tag = "image";            inner = v.value();        }else if (value instanceof FormatMetadata.Xml v) {
+            tag = "xml";            inner = v.value();        }else if (value instanceof FormatMetadata.Text v) {
+            tag = "text";            inner = v.value();        }else if (value instanceof FormatMetadata.Html v) {
+            tag = "html";            inner = v.value();        }else if (value instanceof FormatMetadata.Ocr v) {
+            tag = "ocr";            inner = v.value();        }else if (value instanceof FormatMetadata.Csv v) {
+            tag = "csv";            inner = v.value();        }else if (value instanceof FormatMetadata.Bibtex v) {
+            tag = "bibtex";            inner = v.value();        }else if (value instanceof FormatMetadata.Citation v) {
+            tag = "citation";            inner = v.value();        }else if (value instanceof FormatMetadata.FictionBook v) {
+            tag = "fiction_book";            inner = v.value();        }else if (value instanceof FormatMetadata.Dbf v) {
+            tag = "dbf";            inner = v.value();        }else if (value instanceof FormatMetadata.Jats v) {
+            tag = "jats";            inner = v.value();        }else if (value instanceof FormatMetadata.Epub v) {
+            tag = "epub";            inner = v.value();        }else if (value instanceof FormatMetadata.Pst v) {
+            tag = "pst";            inner = v.value();        }else if (value instanceof FormatMetadata.Code v) {
+            tag = "code";            inner = v.value();        }        else {
+            throw new com.fasterxml.jackson.databind.JsonMappingException(gen,
+                "Unknown FormatMetadata variant: " + value.getClass().getName());
+        }
+
+        gen.writeStartObject();
+        gen.writeStringField("format_type", tag);
+        if (inner != null) {
+            com.fasterxml.jackson.databind.JsonNode tree = MAPPER.valueToTree(inner);
+            if (tree.isObject()) {
+                java.util.Iterator<java.util.Map.Entry<String, com.fasterxml.jackson.databind.JsonNode>> it =
+                    tree.fields();
+                while (it.hasNext()) {
+                    java.util.Map.Entry<String, com.fasterxml.jackson.databind.JsonNode> e = it.next();
+                    gen.writeFieldName(e.getKey());
+                    gen.writeTree(e.getValue());
+                }
+            }
+        }
+        gen.writeEndObject();
+    }
+}

packages/java/dev/kreuzberg/FormattedBlock.java

@@ -0,0 +1,133 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Block-level element in a Djot document.
+ *
+ * Represents structural elements like headings, paragraphs, lists, code blocks, etc.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = FormattedBlock.Builder.class)
+public record FormattedBlock(
+    /**
+     * Type of block element
+     */
+    @JsonProperty("block_type") BlockType blockType,
+    /**
+     * Heading level (1-6) for headings, or nesting level for lists
+     */
+    @Nullable @JsonProperty("level") Long level,
+    /**
+     * Inline content within the block
+     */
+    @JsonProperty("inline_content") List<InlineElement> inlineContent,
+    /**
+     * Element attributes (classes, IDs, key-value pairs)
+     */
+    @Nullable @JsonProperty("attributes") String attributes,
+    /**
+     * Language identifier for code blocks
+     */
+    @Nullable @JsonProperty("language") String language,
+    /**
+     * Raw code content for code blocks
+     */
+    @Nullable @JsonProperty("code") String code,
+    /**
+     * Nested blocks for containers (blockquotes, list items, divs)
+     */
+    @Nullable @JsonProperty("children") List<FormattedBlock> children
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("block_type")
+private BlockType blockType = null;
+private Long level = null;
+        @JsonProperty("inline_content")
+private List<InlineElement> inlineContent = List.of();
+private String attributes = null;
+private String language = null;
+private String code = null;
+private List<FormattedBlock> children = null;
+
+        /** Sets the blockType field. */
+        @JsonProperty("block_type")
+        public Builder withBlockType(final BlockType value) {
+            this.blockType = value;
+            return this;
+        }
+
+        /** Sets the level field. */
+        @JsonProperty("level")
+        public Builder withLevel(final @Nullable long value) {
+            this.level = value;
+            return this;
+        }
+
+        /** Sets the inlineContent field. */
+        @JsonProperty("inline_content")
+        public Builder withInlineContent(final List<InlineElement> value) {
+            this.inlineContent = value;
+            return this;
+        }
+
+        /** Sets the attributes field. */
+        @JsonProperty("attributes")
+        public Builder withAttributes(final @Nullable String value) {
+            this.attributes = value;
+            return this;
+        }
+
+        /** Sets the language field. */
+        @JsonProperty("language")
+        public Builder withLanguage(final @Nullable String value) {
+            this.language = value;
+            return this;
+        }
+
+        /** Sets the code field. */
+        @JsonProperty("code")
+        public Builder withCode(final @Nullable String value) {
+            this.code = value;
+            return this;
+        }
+
+        /** Sets the children field. */
+        @JsonProperty("children")
+        public Builder withChildren(final @Nullable List<FormattedBlock> value) {
+            this.children = value;
+            return this;
+        }
+
+        /** Builds the FormattedBlock instance. */
+        public FormattedBlock build() {
+            return new FormattedBlock(
+                blockType,
+                level,
+                inlineContent,
+                attributes,
+                language,
+                code,
+                children
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/GridCell.java

@@ -0,0 +1,131 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Individual grid cell with position and span metadata.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = GridCell.Builder.class)
+public record GridCell(
+    /**
+     * Cell text content.
+     */
+    @JsonProperty("content") String content,
+    /**
+     * Zero-indexed row position.
+     */
+    @JsonProperty("row") int row,
+    /**
+     * Zero-indexed column position.
+     */
+    @JsonProperty("col") int col,
+    /**
+     * Number of rows this cell spans.
+     */
+    @Nullable @JsonProperty("row_span") Integer rowSpan,
+    /**
+     * Number of columns this cell spans.
+     */
+    @Nullable @JsonProperty("col_span") Integer colSpan,
+    /**
+     * Whether this is a header cell.
+     */
+    @Nullable @JsonProperty("is_header") Boolean isHeader,
+    /**
+     * Bounding box for this cell (if available).
+     */
+    @Nullable @JsonProperty("bbox") BoundingBox bbox
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String content = "";
+private int row = 0;
+private int col = 0;
+        @JsonProperty("row_span")
+private Integer rowSpan = null;
+        @JsonProperty("col_span")
+private Integer colSpan = null;
+        @JsonProperty("is_header")
+private Boolean isHeader = null;
+private BoundingBox bbox = null;
+
+        /** Sets the content field. */
+        @JsonProperty("content")
+        public Builder withContent(final String value) {
+            this.content = value;
+            return this;
+        }
+
+        /** Sets the row field. */
+        @JsonProperty("row")
+        public Builder withRow(final int value) {
+            this.row = value;
+            return this;
+        }
+
+        /** Sets the col field. */
+        @JsonProperty("col")
+        public Builder withCol(final int value) {
+            this.col = value;
+            return this;
+        }
+
+        /** Sets the rowSpan field. */
+        @JsonProperty("row_span")
+        public Builder withRowSpan(final @Nullable Integer value) {
+            this.rowSpan = value;
+            return this;
+        }
+
+        /** Sets the colSpan field. */
+        @JsonProperty("col_span")
+        public Builder withColSpan(final @Nullable Integer value) {
+            this.colSpan = value;
+            return this;
+        }
+
+        /** Sets the isHeader field. */
+        @JsonProperty("is_header")
+        public Builder withIsHeader(final @Nullable Boolean value) {
+            this.isHeader = value;
+            return this;
+        }
+
+        /** Sets the bbox field. */
+        @JsonProperty("bbox")
+        public Builder withBbox(final @Nullable BoundingBox value) {
+            this.bbox = value;
+            return this;
+        }
+
+        /** Builds the GridCell instance. */
+        public GridCell build() {
+            return new GridCell(
+                content,
+                row,
+                col,
+                rowSpan,
+                colSpan,
+                isHeader,
+                bbox
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/HeaderMetadata.java

@@ -0,0 +1,103 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Header/heading element metadata.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = HeaderMetadata.Builder.class)
+public record HeaderMetadata(
+    /**
+     * Header level: 1 (h1) through 6 (h6)
+     */
+    @JsonProperty("level") byte level,
+    /**
+     * Normalized text content of the header
+     */
+    @JsonProperty("text") String text,
+    /**
+     * HTML id attribute if present
+     */
+    @Nullable @JsonProperty("id") String id,
+    /**
+     * Document tree depth at the header element
+     */
+    @JsonProperty("depth") int depth,
+    /**
+     * Byte offset in original HTML document
+     */
+    @JsonProperty("html_offset") int htmlOffset
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private byte level = 0;
+private String text = "";
+private String id = null;
+private int depth = 0;
+        @JsonProperty("html_offset")
+private int htmlOffset = 0;
+
+        /** Sets the level field. */
+        @JsonProperty("level")
+        public Builder withLevel(final byte value) {
+            this.level = value;
+            return this;
+        }
+
+        /** Sets the text field. */
+        @JsonProperty("text")
+        public Builder withText(final String value) {
+            this.text = value;
+            return this;
+        }
+
+        /** Sets the id field. */
+        @JsonProperty("id")
+        public Builder withId(final @Nullable String value) {
+            this.id = value;
+            return this;
+        }
+
+        /** Sets the depth field. */
+        @JsonProperty("depth")
+        public Builder withDepth(final int value) {
+            this.depth = value;
+            return this;
+        }
+
+        /** Sets the htmlOffset field. */
+        @JsonProperty("html_offset")
+        public Builder withHtmlOffset(final int value) {
+            this.htmlOffset = value;
+            return this;
+        }
+
+        /** Builds the HeaderMetadata instance. */
+        public HeaderMetadata build() {
+            return new HeaderMetadata(
+                level,
+                text,
+                id,
+                depth,
+                htmlOffset
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/HeadingContext.java

@@ -0,0 +1,47 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * Heading context for a chunk within a Markdown document.
+ *
+ * Contains the heading hierarchy from document root to this chunk's section.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = HeadingContext.Builder.class)
+public record HeadingContext(@JsonProperty("headings") List<HeadingLevel> headings) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private List<HeadingLevel> headings = List.of();
+
+        /** Sets the headings field. */
+        @JsonProperty("headings")
+        public Builder withHeadings(final List<HeadingLevel> value) {
+            this.headings = value;
+            return this;
+        }
+
+        /** Builds the HeadingContext instance. */
+        public HeadingContext build() {
+            return new HeadingContext(
+                headings
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/HeadingLevel.java

@@ -0,0 +1,53 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * A single heading in the hierarchy.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = HeadingLevel.Builder.class)
+public record HeadingLevel(@JsonProperty("level") byte level, @JsonProperty("text") String text) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private byte level = 0;
+private String text = "";
+
+        /** Sets the level field. */
+        @JsonProperty("level")
+        public Builder withLevel(final byte value) {
+            this.level = value;
+            return this;
+        }
+
+        /** Sets the text field. */
+        @JsonProperty("text")
+        public Builder withText(final String value) {
+            this.text = value;
+            return this;
+        }
+
+        /** Builds the HeadingLevel instance. */
+        public HeadingLevel build() {
+            return new HeadingLevel(
+                level,
+                text
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/HierarchicalBlock.java

@@ -0,0 +1,105 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * A text block with hierarchy level assignment.
+ *
+ * Represents a block of text with semantic heading information extracted from
+ * font size clustering and hierarchical analysis.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = HierarchicalBlock.Builder.class)
+public record HierarchicalBlock(
+    /**
+     * The text content of this block
+     */
+    @JsonProperty("text") String text,
+    /**
+     * The font size of the text in this block
+     */
+    @JsonProperty("font_size") float fontSize,
+    /**
+     * The hierarchy level of this block (H1-H6 or Body)
+     *
+     * Levels correspond to HTML heading tags:
+     * - "h1": Top-level heading
+     * - "h2": Secondary heading
+     * - "h3": Tertiary heading
+     * - "h4": Quaternary heading
+     * - "h5": Quinary heading
+     * - "h6": Senary heading
+     * - "body": Body text (no heading level)
+     */
+    @JsonProperty("level") String level,
+    /**
+     * Bounding box information for the block
+     *
+     * Contains coordinates as (left, top, right, bottom) in PDF units.
+     */
+    @Nullable @JsonProperty("bbox") List<Float> bbox
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String text = "";
+        @JsonProperty("font_size")
+private float fontSize = 0.0f;
+private String level = "";
+private List<Float> bbox = null;
+
+        /** Sets the text field. */
+        @JsonProperty("text")
+        public Builder withText(final String value) {
+            this.text = value;
+            return this;
+        }
+
+        /** Sets the fontSize field. */
+        @JsonProperty("font_size")
+        public Builder withFontSize(final float value) {
+            this.fontSize = value;
+            return this;
+        }
+
+        /** Sets the level field. */
+        @JsonProperty("level")
+        public Builder withLevel(final String value) {
+            this.level = value;
+            return this;
+        }
+
+        /** Sets the bbox field. */
+        @JsonProperty("bbox")
+        public Builder withBbox(final @Nullable List<Float> value) {
+            this.bbox = value;
+            return this;
+        }
+
+        /** Builds the HierarchicalBlock instance. */
+        public HierarchicalBlock build() {
+            return new HierarchicalBlock(
+                text,
+                fontSize,
+                level,
+                bbox
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/HierarchyConfig.java

@@ -0,0 +1,109 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Hierarchy extraction configuration for PDF text structure analysis.
+ *
+ * Enables extraction of document hierarchy levels (H1-H6) based on font size
+ * clustering and semantic analysis. When enabled, hierarchical blocks are
+ * included in page content.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = HierarchyConfig.Builder.class)
+public record HierarchyConfig(
+    /**
+     * Enable hierarchy extraction
+     */
+    @Nullable @JsonProperty("enabled") Boolean enabled,
+    /**
+     * Number of font size clusters to use for hierarchy levels (1-7)
+     *
+     * Default: 6, which provides H1-H6 heading levels with body text.
+     * Larger values create more fine-grained hierarchy levels.
+     */
+    @Nullable @JsonProperty("k_clusters") Long kClusters,
+    /**
+     * Include bounding box information in hierarchy blocks
+     */
+    @Nullable @JsonProperty("include_bbox") Boolean includeBbox,
+    /**
+     * OCR coverage threshold for smart OCR triggering (0.0-1.0)
+     *
+     * Determines when OCR should be triggered based on text block coverage.
+     * OCR is triggered when text blocks cover less than this fraction of the page.
+     * Default: 0.5 (trigger OCR if less than 50% of page has text)
+     */
+    @Nullable @JsonProperty("ocr_coverage_threshold") Float ocrCoverageThreshold
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+    public HierarchyConfig{
+        if (kClusters == null) kClusters = 3L;
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private Boolean enabled = null;
+        @JsonProperty("k_clusters")
+private Long kClusters = null;
+        @JsonProperty("include_bbox")
+private Boolean includeBbox = null;
+        @JsonProperty("ocr_coverage_threshold")
+private Float ocrCoverageThreshold = null;
+
+        /** Sets the enabled field. */
+        @JsonProperty("enabled")
+        public Builder withEnabled(final @Nullable Boolean value) {
+            this.enabled = value;
+            return this;
+        }
+
+        /** Sets the kClusters field. */
+        @JsonProperty("k_clusters")
+        public Builder withKClusters(final @Nullable Long value) {
+            this.kClusters = value;
+            return this;
+        }
+
+        /** Sets the includeBbox field. */
+        @JsonProperty("include_bbox")
+        public Builder withIncludeBbox(final @Nullable Boolean value) {
+            this.includeBbox = value;
+            return this;
+        }
+
+        /** Sets the ocrCoverageThreshold field. */
+        @JsonProperty("ocr_coverage_threshold")
+        public Builder withOcrCoverageThreshold(final @Nullable Float value) {
+            this.ocrCoverageThreshold = value;
+            return this;
+        }
+
+        /** Builds the HierarchyConfig instance. */
+        public HierarchyConfig build() {
+            return new HierarchyConfig(
+                enabled,
+                kClusters,
+                includeBbox,
+                ocrCoverageThreshold
+            );
+        }
+    }
+    // CPD-ON
+    public static HierarchyConfig defaultInstance() {
+        throw new UnsupportedOperationException("defaultInstance is not yet bridged via JNI; use the Builder instead.");
+    }
+}

packages/java/dev/kreuzberg/HtmlMetadata.java

@@ -0,0 +1,247 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import java.util.Map;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * HTML metadata extracted from HTML documents.
+ *
+ * Includes document-level metadata, Open Graph data, Twitter Card metadata,
+ * and extracted structural elements (headers, links, images, structured data).
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = HtmlMetadata.Builder.class)
+public record HtmlMetadata(
+    /**
+     * Document title from {@code <title>} tag
+     */
+    @Nullable @JsonProperty("title") String title,
+    /**
+     * Document description from {@code <meta name="description">} tag
+     */
+    @Nullable @JsonProperty("description") String description,
+    /**
+     * Document keywords from {@code <meta name="keywords">} tag, split on commas
+     */
+    @Nullable @JsonProperty("keywords") List<String> keywords,
+    /**
+     * Document author from {@code &lt;meta name="author"&gt;} tag
+     */
+    @Nullable @JsonProperty("author") String author,
+    /**
+     * Canonical URL from {@code &lt;link rel="canonical"&gt;} tag
+     */
+    @Nullable @JsonProperty("canonical_url") String canonicalUrl,
+    /**
+     * Base URL from {@code &lt;base href=""&gt;} tag for resolving relative URLs
+     */
+    @Nullable @JsonProperty("base_href") String baseHref,
+    /**
+     * Document language from {@code lang} attribute
+     */
+    @Nullable @JsonProperty("language") String language,
+    /**
+     * Document text direction from {@code dir} attribute
+     */
+    @Nullable @JsonProperty("text_direction") TextDirection textDirection,
+    /**
+     * Open Graph metadata (og:* properties) for social media
+     * Keys like "title", "description", "image", "url", etc.
+     */
+    @Nullable @JsonProperty("open_graph") Map<String, String> openGraph,
+    /**
+     * Twitter Card metadata (twitter:* properties)
+     * Keys like "card", "site", "creator", "title", "description", "image", etc.
+     */
+    @Nullable @JsonProperty("twitter_card") Map<String, String> twitterCard,
+    /**
+     * Additional meta tags not covered by specific fields
+     * Keys are meta name/property attributes, values are content
+     */
+    @Nullable @JsonProperty("meta_tags") Map<String, String> metaTags,
+    /**
+     * Extracted header elements with hierarchy
+     */
+    @Nullable @JsonProperty("headers") List<HeaderMetadata> headers,
+    /**
+     * Extracted hyperlinks with type classification
+     */
+    @Nullable @JsonProperty("links") List<LinkMetadata> links,
+    /**
+     * Extracted images with source and dimensions
+     */
+    @Nullable @JsonProperty("images") List<ImageMetadataType> images,
+    /**
+     * Extracted structured data blocks
+     */
+    @Nullable @JsonProperty("structured_data") List<StructuredData> structuredData
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String title = null;
+private String description = null;
+private List<String> keywords = null;
+private String author = null;
+        @JsonProperty("canonical_url")
+private String canonicalUrl = null;
+        @JsonProperty("base_href")
+private String baseHref = null;
+private String language = null;
+        @JsonProperty("text_direction")
+private TextDirection textDirection = null;
+        @JsonProperty("open_graph")
+private Map<String, String> openGraph = null;
+        @JsonProperty("twitter_card")
+private Map<String, String> twitterCard = null;
+        @JsonProperty("meta_tags")
+private Map<String, String> metaTags = null;
+private List<HeaderMetadata> headers = null;
+private List<LinkMetadata> links = null;
+private List<ImageMetadataType> images = null;
+        @JsonProperty("structured_data")
+private List<StructuredData> structuredData = null;
+
+        /** Sets the title field. */
+        @JsonProperty("title")
+        public Builder withTitle(final @Nullable String value) {
+            this.title = value;
+            return this;
+        }
+
+        /** Sets the description field. */
+        @JsonProperty("description")
+        public Builder withDescription(final @Nullable String value) {
+            this.description = value;
+            return this;
+        }
+
+        /** Sets the keywords field. */
+        @JsonProperty("keywords")
+        public Builder withKeywords(final @Nullable List<String> value) {
+            this.keywords = value;
+            return this;
+        }
+
+        /** Sets the author field. */
+        @JsonProperty("author")
+        public Builder withAuthor(final @Nullable String value) {
+            this.author = value;
+            return this;
+        }
+
+        /** Sets the canonicalUrl field. */
+        @JsonProperty("canonical_url")
+        public Builder withCanonicalUrl(final @Nullable String value) {
+            this.canonicalUrl = value;
+            return this;
+        }
+
+        /** Sets the baseHref field. */
+        @JsonProperty("base_href")
+        public Builder withBaseHref(final @Nullable String value) {
+            this.baseHref = value;
+            return this;
+        }
+
+        /** Sets the language field. */
+        @JsonProperty("language")
+        public Builder withLanguage(final @Nullable String value) {
+            this.language = value;
+            return this;
+        }
+
+        /** Sets the textDirection field. */
+        @JsonProperty("text_direction")
+        public Builder withTextDirection(final @Nullable TextDirection value) {
+            this.textDirection = value;
+            return this;
+        }
+
+        /** Sets the openGraph field. */
+        @JsonProperty("open_graph")
+        public Builder withOpenGraph(final @Nullable Map<String, String> value) {
+            this.openGraph = value;
+            return this;
+        }
+
+        /** Sets the twitterCard field. */
+        @JsonProperty("twitter_card")
+        public Builder withTwitterCard(final @Nullable Map<String, String> value) {
+            this.twitterCard = value;
+            return this;
+        }
+
+        /** Sets the metaTags field. */
+        @JsonProperty("meta_tags")
+        public Builder withMetaTags(final @Nullable Map<String, String> value) {
+            this.metaTags = value;
+            return this;
+        }
+
+        /** Sets the headers field. */
+        @JsonProperty("headers")
+        public Builder withHeaders(final @Nullable List<HeaderMetadata> value) {
+            this.headers = value;
+            return this;
+        }
+
+        /** Sets the links field. */
+        @JsonProperty("links")
+        public Builder withLinks(final @Nullable List<LinkMetadata> value) {
+            this.links = value;
+            return this;
+        }
+
+        /** Sets the images field. */
+        @JsonProperty("images")
+        public Builder withImages(final @Nullable List<ImageMetadataType> value) {
+            this.images = value;
+            return this;
+        }
+
+        /** Sets the structuredData field. */
+        @JsonProperty("structured_data")
+        public Builder withStructuredData(final @Nullable List<StructuredData> value) {
+            this.structuredData = value;
+            return this;
+        }
+
+        /** Builds the HtmlMetadata instance. */
+        public HtmlMetadata build() {
+            return new HtmlMetadata(
+                title,
+                description,
+                keywords,
+                author,
+                canonicalUrl,
+                baseHref,
+                language,
+                textDirection,
+                openGraph,
+                twitterCard,
+                metaTags,
+                headers,
+                links,
+                images,
+                structuredData
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/HtmlOutputConfig.java

@@ -0,0 +1,122 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Configuration for styled HTML output.
+ *
+ * When set on ExtractionConfig.html_output alongside
+ * {@code output_format = OutputFormat.Html}, the pipeline builds a
+ * StyledHtmlRenderer(crate.rendering.StyledHtmlRenderer) instead of
+ * the plain comrak-based renderer.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = HtmlOutputConfig.Builder.class)
+public record HtmlOutputConfig(
+    /**
+     * Inline CSS string injected into the output after the theme stylesheet.
+     * Concatenated after {@code css_file} content when both are set.
+     */
+    @Nullable @JsonProperty("css") String css,
+    /**
+     * Path to a CSS file loaded once at renderer construction time.
+     * Concatenated before {@code css} when both are set.
+     */
+    @JsonProperty("css_file") java.nio.file.@Nullable Path cssFile,
+    /**
+     * Built-in colour/typography theme. Default: HtmlTheme.Unstyled.
+     */
+    @Nullable @JsonProperty("theme") HtmlTheme theme,
+    /**
+     * CSS class prefix applied to every emitted class name.
+     *
+     * Default: {@code "kb-"}. Change this if your host application already uses
+     * classes that start with {@code kb-}.
+     */
+    @Nullable @JsonProperty("class_prefix") String classPrefix,
+    /**
+     * When {@code true} (default), write the resolved CSS into a {@code <style>} block
+     * immediately after the opening {@code <div class="{prefix}doc">}.
+     *
+     * Set to {@code false} to emit only the structural markup and wire up your
+     * own stylesheet targeting the {@code kb-*} class names.
+     */
+    @Nullable @JsonProperty("embed_css") Boolean embedCss
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String css = null;
+        @JsonProperty("css_file")
+private java.nio.file.Path cssFile = null;
+        @Nullable private HtmlTheme theme = HtmlTheme.Unstyled;
+        @JsonProperty("class_prefix")
+private String classPrefix = null;
+        @JsonProperty("embed_css")
+private Boolean embedCss = null;
+
+        /** Sets the css field. */
+        @JsonProperty("css")
+        public Builder withCss(final @Nullable String value) {
+            this.css = value;
+            return this;
+        }
+
+        /** Sets the cssFile field. */
+        @JsonProperty("css_file")
+        public Builder withCssFile(final java.nio.file.@Nullable Path value) {
+            this.cssFile = value;
+            return this;
+        }
+
+        /** Sets the theme field. */
+        @JsonProperty("theme")
+        public Builder withTheme(final @Nullable HtmlTheme value) {
+            this.theme = value;
+            return this;
+        }
+
+        /** Sets the classPrefix field. */
+        @JsonProperty("class_prefix")
+        public Builder withClassPrefix(final @Nullable String value) {
+            this.classPrefix = value;
+            return this;
+        }
+
+        /** Sets the embedCss field. */
+        @JsonProperty("embed_css")
+        public Builder withEmbedCss(final @Nullable Boolean value) {
+            this.embedCss = value;
+            return this;
+        }
+
+        /** Builds the HtmlOutputConfig instance. */
+        public HtmlOutputConfig build() {
+            return new HtmlOutputConfig(
+                css,
+                cssFile,
+                theme,
+                classPrefix,
+                embedCss
+            );
+        }
+    }
+    // CPD-ON
+    public static HtmlOutputConfig defaultInstance() {
+        throw new UnsupportedOperationException("defaultInstance is not yet bridged via JNI; use the Builder instead.");
+    }
+}

packages/java/dev/kreuzberg/HtmlTheme.java

@@ -0,0 +1,57 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonCreator;
+import com.fasterxml.jackson.annotation.JsonValue;
+
+/**
+ * Built-in HTML theme selection.
+ */
+public enum HtmlTheme {
+    /** Sensible defaults: system font stack, neutral colours, readable line */
+    Default("default"),
+    /** GitHub Markdown-inspired palette and spacing. */
+    GitHub("github"),
+    /** Dark background, light text. */
+    Dark("dark"),
+    /** Minimal light theme with generous whitespace. */
+    Light("light"),
+    /**
+     * No built-in stylesheet emitted. CSS custom properties are still defined
+     */
+    Unstyled("unstyled");
+
+    /** The string value. */
+    private final String value;
+
+    HtmlTheme(final String value) {
+        this.value = value;
+    }
+
+    /** Returns the string value. */
+    @JsonValue
+    public String getValue() {
+        return value;
+    }
+
+    /** Creates an instance from a string value. */
+    @JsonCreator
+    public static HtmlTheme fromValue(final String value) {
+        for (HtmlTheme e : values()) {
+            if (e.value.equalsIgnoreCase(value)) {
+                return e;
+            }
+        }
+        throw new IllegalArgumentException("Unknown value: " + value);
+    }
+
+    /** Returns the wire-format string value (matches JSON serialization). */
+    @Override
+    public String toString() {
+        return value;
+    }
+}

packages/java/dev/kreuzberg/IDocumentExtractor.java

@@ -0,0 +1,34 @@
+package dev.kreuzberg;
+
+import java.util.List;
+
+/**
+ * Bridge interface for the DocumentExtractor plugin system.
+ *
+ * Implementations are wrapped by DocumentExtractorBridge and exposed to the native
+ * runtime through Panama FFM upcall stubs.
+ */
+public interface IDocumentExtractor {
+
+    /** Plugin name (used for registry keying). */
+    String name();
+
+    /** Plugin version. */
+    String version();
+
+    /** Initialize the plugin. */
+    default void initialize() throws Exception {}
+
+    /** Shut down the plugin. */
+    default void shutdown() throws Exception {}
+
+/** extract_bytes. */    String extract_bytes(byte[] content, String mime_type, ExtractionConfig config) throws Exception;
+
+/** extract_file. */    String extract_file(java.nio.file.Path path, String mime_type, ExtractionConfig config) throws Exception;
+
+/** supported_mime_types. */    List<String> supported_mime_types() throws Exception;
+
+/** priority. */    int priority() throws Exception;
+
+/** can_handle. */    boolean can_handle(java.nio.file.Path _path, String _mime_type) throws Exception;
+}

packages/java/dev/kreuzberg/IEmbeddingBackend.java

@@ -0,0 +1,28 @@
+package dev.kreuzberg;
+
+import java.util.List;
+
+/**
+ * Bridge interface for the EmbeddingBackend plugin system.
+ *
+ * Implementations are wrapped by EmbeddingBackendBridge and exposed to the native
+ * runtime through Panama FFM upcall stubs.
+ */
+public interface IEmbeddingBackend {
+
+    /** Plugin name (used for registry keying). */
+    String name();
+
+    /** Plugin version. */
+    String version();
+
+    /** Initialize the plugin. */
+    default void initialize() throws Exception {}
+
+    /** Shut down the plugin. */
+    default void shutdown() throws Exception {}
+
+/** dimensions. */    long dimensions() throws Exception;
+
+/** embed. */    List<List<Float>> embed(List<String> texts) throws Exception;
+}

packages/java/dev/kreuzberg/IOcrBackend.java

@@ -0,0 +1,40 @@
+package dev.kreuzberg;
+
+import java.util.List;
+
+/**
+ * Bridge interface for the OcrBackend plugin system.
+ *
+ * Implementations are wrapped by OcrBackendBridge and exposed to the native
+ * runtime through Panama FFM upcall stubs.
+ */
+public interface IOcrBackend {
+
+    /** Plugin name (used for registry keying). */
+    String name();
+
+    /** Plugin version. */
+    String version();
+
+    /** Initialize the plugin. */
+    default void initialize() throws Exception {}
+
+    /** Shut down the plugin. */
+    default void shutdown() throws Exception {}
+
+/** process_image. */    ExtractionResult process_image(byte[] image_bytes, OcrConfig config) throws Exception;
+
+/** process_image_file. */    ExtractionResult process_image_file(java.nio.file.Path path, OcrConfig config) throws Exception;
+
+/** supports_language. */    boolean supports_language(String lang) throws Exception;
+
+/** backend_type. */    String backend_type() throws Exception;
+
+/** supported_languages. */    List<String> supported_languages() throws Exception;
+
+/** supports_table_detection. */    boolean supports_table_detection() throws Exception;
+
+/** supports_document_processing. */    boolean supports_document_processing() throws Exception;
+
+/** process_document. */    ExtractionResult process_document(java.nio.file.Path _path, OcrConfig _config) throws Exception;
+}

packages/java/dev/kreuzberg/IPostProcessor.java

@@ -0,0 +1,32 @@
+package dev.kreuzberg;
+
+/**
+ * Bridge interface for the PostProcessor plugin system.
+ *
+ * Implementations are wrapped by PostProcessorBridge and exposed to the native
+ * runtime through Panama FFM upcall stubs.
+ */
+public interface IPostProcessor {
+
+    /** Plugin name (used for registry keying). */
+    String name();
+
+    /** Plugin version. */
+    String version();
+
+    /** Initialize the plugin. */
+    default void initialize() throws Exception {}
+
+    /** Shut down the plugin. */
+    default void shutdown() throws Exception {}
+
+/** process. */    void process(ExtractionResult result, ExtractionConfig config) throws Exception;
+
+/** processing_stage. */    String processing_stage() throws Exception;
+
+/** should_process. */    boolean should_process(ExtractionResult _result, ExtractionConfig _config) throws Exception;
+
+/** estimated_duration_ms. */    long estimated_duration_ms(ExtractionResult _result) throws Exception;
+
+/** priority. */    int priority() throws Exception;
+}

packages/java/dev/kreuzberg/IRenderer.java

@@ -0,0 +1,24 @@
+package dev.kreuzberg;
+
+/**
+ * Bridge interface for the Renderer plugin system.
+ *
+ * Implementations are wrapped by RendererBridge and exposed to the native
+ * runtime through Panama FFM upcall stubs.
+ */
+public interface IRenderer {
+
+    /** Plugin name (used for registry keying). */
+    String name();
+
+    /** Plugin version. */
+    String version();
+
+    /** Initialize the plugin. */
+    default void initialize() throws Exception {}
+
+    /** Shut down the plugin. */
+    default void shutdown() throws Exception {}
+
+/** render. */    String render(String doc) throws Exception;
+}

packages/java/dev/kreuzberg/IValidator.java

@@ -0,0 +1,28 @@
+package dev.kreuzberg;
+
+/**
+ * Bridge interface for the Validator plugin system.
+ *
+ * Implementations are wrapped by ValidatorBridge and exposed to the native
+ * runtime through Panama FFM upcall stubs.
+ */
+public interface IValidator {
+
+    /** Plugin name (used for registry keying). */
+    String name();
+
+    /** Plugin version. */
+    String version();
+
+    /** Initialize the plugin. */
+    default void initialize() throws Exception {}
+
+    /** Shut down the plugin. */
+    default void shutdown() throws Exception {}
+
+/** validate. */    void validate(ExtractionResult result, ExtractionConfig config) throws Exception;
+
+/** should_validate. */    boolean should_validate(ExtractionResult _result, ExtractionConfig _config) throws Exception;
+
+/** priority. */    int priority() throws Exception;
+}

packages/java/dev/kreuzberg/ImageExtractionConfig.java

@@ -0,0 +1,255 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Image extraction configuration.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ImageExtractionConfig.Builder.class)
+public record ImageExtractionConfig(
+    /**
+     * Extract images from documents
+     */
+    @Nullable @JsonProperty("extract_images") Boolean extractImages,
+    /**
+     * Target DPI for image normalization
+     */
+    @Nullable @JsonProperty("target_dpi") Integer targetDpi,
+    /**
+     * Maximum dimension for images (width or height)
+     */
+    @Nullable @JsonProperty("max_image_dimension") Integer maxImageDimension,
+    /**
+     * Whether to inject image reference placeholders into markdown output.
+     * When {@code true} (default), image references like {@code ![Image 1](embedded:p1_i0)}
+     * are appended to the markdown. Set to {@code false} to extract images as data
+     * without polluting the markdown output.
+     */
+    @Nullable @JsonProperty("inject_placeholders") Boolean injectPlaceholders,
+    /**
+     * Automatically adjust DPI based on image content
+     */
+    @Nullable @JsonProperty("auto_adjust_dpi") Boolean autoAdjustDpi,
+    /**
+     * Minimum DPI threshold
+     */
+    @Nullable @JsonProperty("min_dpi") Integer minDpi,
+    /**
+     * Maximum DPI threshold
+     */
+    @Nullable @JsonProperty("max_dpi") Integer maxDpi,
+    /**
+     * Maximum number of image objects to extract per PDF page.
+     *
+     * Some PDFs (e.g. technical diagrams stored as thousands of raster fragments)
+     * can trigger extremely long or indefinite extraction times when every image
+     * object on a dense page is decoded individually via the PDF extractor. Setting this
+     * limit causes kreuzberg to stop collecting individual images once the count
+     * per page reaches the cap and emit a warning instead.
+     *
+     * {@code None} (default) means no limit — all images are extracted.
+     */
+    @Nullable @JsonProperty("max_images_per_page") Integer maxImagesPerPage,
+    /**
+     * When {@code true} (default), extracted images are classified by kind and grouped
+     * into clusters where they appear to belong to one figure.
+     */
+    @Nullable @JsonProperty("classify") Boolean classify,
+    /**
+     * When {@code true}, full-page renders produced during OCR preprocessing are captured
+     * and returned as {@code ImageKind.PageRaster} entries in {@code ExtractionResult.images}.
+     *
+     * **PDF + OCR only.** No rasters are captured for non-PDF inputs or when the
+     * document-level OCR bypass is active (whole-document backend). When OCR is
+     * enabled and this flag is set but the active backend skips per-page rendering,
+     * a {@code ProcessingWarning} is emitted in {@code ExtractionResult.processing_warnings}.
+     *
+     * Defaults to {@code false}. Enable when downstream consumers need page thumbnails
+     * (e.g. citation previews, visual grounding).
+     */
+    @Nullable @JsonProperty("include_page_rasters") Boolean includePageRasters,
+    /**
+     * Run OCR on extracted images and include the recognized text in the document content.
+     *
+     * When {@code true} (default) and {@code ExtractionConfig.ocr} is configured, extracted images
+     * are processed with the configured OCR backend. Set to {@code false} to extract images
+     * without OCR processing, even when OCR is enabled.
+     */
+    @Nullable @JsonProperty("run_ocr_on_images") Boolean runOcrOnImages,
+    /**
+     * When {@code true}, image OCR results are rendered as plain text without the
+     * {@code ![...](...)} markdown placeholder. Only takes effect when {@code run_ocr_on_images}
+     * is also {@code true}.
+     */
+    @Nullable @JsonProperty("ocr_text_only") Boolean ocrTextOnly,
+    /**
+     * When {@code true} and {@code ocr_text_only} is {@code false}, append the OCR text after
+     * the image placeholder in the rendered output.
+     */
+    @Nullable @JsonProperty("append_ocr_text") Boolean appendOcrText
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+    public ImageExtractionConfig{
+        if (targetDpi == null) targetDpi = 300;
+        if (maxImageDimension == null) maxImageDimension = 4096;
+        if (minDpi == null) minDpi = 72;
+        if (maxDpi == null) maxDpi = 600;
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("extract_images")
+private Boolean extractImages = null;
+        @JsonProperty("target_dpi")
+private Integer targetDpi = null;
+        @JsonProperty("max_image_dimension")
+private Integer maxImageDimension = null;
+        @JsonProperty("inject_placeholders")
+private Boolean injectPlaceholders = null;
+        @JsonProperty("auto_adjust_dpi")
+private Boolean autoAdjustDpi = null;
+        @JsonProperty("min_dpi")
+private Integer minDpi = null;
+        @JsonProperty("max_dpi")
+private Integer maxDpi = null;
+        @JsonProperty("max_images_per_page")
+private Integer maxImagesPerPage = null;
+private Boolean classify = null;
+        @JsonProperty("include_page_rasters")
+private Boolean includePageRasters = null;
+        @JsonProperty("run_ocr_on_images")
+private Boolean runOcrOnImages = null;
+        @JsonProperty("ocr_text_only")
+private Boolean ocrTextOnly = null;
+        @JsonProperty("append_ocr_text")
+private Boolean appendOcrText = null;
+
+        /** Sets the extractImages field. */
+        @JsonProperty("extract_images")
+        public Builder withExtractImages(final @Nullable Boolean value) {
+            this.extractImages = value;
+            return this;
+        }
+
+        /** Sets the targetDpi field. */
+        @JsonProperty("target_dpi")
+        public Builder withTargetDpi(final @Nullable Integer value) {
+            this.targetDpi = value;
+            return this;
+        }
+
+        /** Sets the maxImageDimension field. */
+        @JsonProperty("max_image_dimension")
+        public Builder withMaxImageDimension(final @Nullable Integer value) {
+            this.maxImageDimension = value;
+            return this;
+        }
+
+        /** Sets the injectPlaceholders field. */
+        @JsonProperty("inject_placeholders")
+        public Builder withInjectPlaceholders(final @Nullable Boolean value) {
+            this.injectPlaceholders = value;
+            return this;
+        }
+
+        /** Sets the autoAdjustDpi field. */
+        @JsonProperty("auto_adjust_dpi")
+        public Builder withAutoAdjustDpi(final @Nullable Boolean value) {
+            this.autoAdjustDpi = value;
+            return this;
+        }
+
+        /** Sets the minDpi field. */
+        @JsonProperty("min_dpi")
+        public Builder withMinDpi(final @Nullable Integer value) {
+            this.minDpi = value;
+            return this;
+        }
+
+        /** Sets the maxDpi field. */
+        @JsonProperty("max_dpi")
+        public Builder withMaxDpi(final @Nullable Integer value) {
+            this.maxDpi = value;
+            return this;
+        }
+
+        /** Sets the maxImagesPerPage field. */
+        @JsonProperty("max_images_per_page")
+        public Builder withMaxImagesPerPage(final @Nullable Integer value) {
+            this.maxImagesPerPage = value;
+            return this;
+        }
+
+        /** Sets the classify field. */
+        @JsonProperty("classify")
+        public Builder withClassify(final @Nullable Boolean value) {
+            this.classify = value;
+            return this;
+        }
+
+        /** Sets the includePageRasters field. */
+        @JsonProperty("include_page_rasters")
+        public Builder withIncludePageRasters(final @Nullable Boolean value) {
+            this.includePageRasters = value;
+            return this;
+        }
+
+        /** Sets the runOcrOnImages field. */
+        @JsonProperty("run_ocr_on_images")
+        public Builder withRunOcrOnImages(final @Nullable Boolean value) {
+            this.runOcrOnImages = value;
+            return this;
+        }
+
+        /** Sets the ocrTextOnly field. */
+        @JsonProperty("ocr_text_only")
+        public Builder withOcrTextOnly(final @Nullable Boolean value) {
+            this.ocrTextOnly = value;
+            return this;
+        }
+
+        /** Sets the appendOcrText field. */
+        @JsonProperty("append_ocr_text")
+        public Builder withAppendOcrText(final @Nullable Boolean value) {
+            this.appendOcrText = value;
+            return this;
+        }
+
+        /** Builds the ImageExtractionConfig instance. */
+        public ImageExtractionConfig build() {
+            return new ImageExtractionConfig(
+                extractImages,
+                targetDpi,
+                maxImageDimension,
+                injectPlaceholders,
+                autoAdjustDpi,
+                minDpi,
+                maxDpi,
+                maxImagesPerPage,
+                classify,
+                includePageRasters,
+                runOcrOnImages,
+                ocrTextOnly,
+                appendOcrText
+            );
+        }
+    }
+    // CPD-ON
+    public static ImageExtractionConfig defaultInstance() {
+        throw new UnsupportedOperationException("defaultInstance is not yet bridged via JNI; use the Builder instead.");
+    }
+}

packages/java/dev/kreuzberg/ImageKind.java

@@ -0,0 +1,71 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonCreator;
+import com.fasterxml.jackson.annotation.JsonValue;
+
+/**
+ * Heuristic classification of what an image likely depicts.
+ */
+public enum ImageKind {
+    /** Photographic image (natural scene, photograph) */
+    Photograph("photograph"),
+    /** Technical or schematic diagram */
+    Diagram("diagram"),
+    /** Chart, graph, or plot */
+    Chart("chart"),
+    /** Freehand or technical drawing */
+    Drawing("drawing"),
+    /** Text-heavy image (scanned text, document) */
+    TextBlock("text_block"),
+    /** Decorative element or border */
+    Decoration("decoration"),
+    /** Logo or brand mark */
+    Logo("logo"),
+    /** Small icon */
+    Icon("icon"),
+    /** Fragment of a larger tiled image (tile of a technical drawing) */
+    TileFragment("tile_fragment"),
+    /** Mask or transparency map */
+    Mask("mask"),
+    /**
+     * Full-page render produced during OCR preprocessing; used as a citation thumbnail.
+     */
+    PageRaster("page_raster"),
+    /** Could not classify with reasonable confidence */
+    Unknown("unknown");
+
+    /** The string value. */
+    private final String value;
+
+    ImageKind(final String value) {
+        this.value = value;
+    }
+
+    /** Returns the string value. */
+    @JsonValue
+    public String getValue() {
+        return value;
+    }
+
+    /** Creates an instance from a string value. */
+    @JsonCreator
+    public static ImageKind fromValue(final String value) {
+        for (ImageKind e : values()) {
+            if (e.value.equalsIgnoreCase(value)) {
+                return e;
+            }
+        }
+        throw new IllegalArgumentException("Unknown value: " + value);
+    }
+
+    /** Returns the wire-format string value (matches JSON serialization). */
+    @Override
+    public String toString() {
+        return value;
+    }
+}

packages/java/dev/kreuzberg/ImageMetadata.java

@@ -0,0 +1,91 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.Map;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * Image metadata extracted from image files.
+ *
+ * Includes dimensions, format, and EXIF data.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ImageMetadata.Builder.class)
+public record ImageMetadata(
+    /**
+     * Image width in pixels
+     */
+    @JsonProperty("width") int width,
+    /**
+     * Image height in pixels
+     */
+    @JsonProperty("height") int height,
+    /**
+     * Image format (e.g., "PNG", "JPEG", "TIFF")
+     */
+    @JsonProperty("format") String format,
+    /**
+     * EXIF metadata tags
+     */
+    @JsonProperty("exif") Map<String, String> exif
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private int width = 0;
+private int height = 0;
+private String format = "";
+private Map<String, String> exif = Map.of();
+
+        /** Sets the width field. */
+        @JsonProperty("width")
+        public Builder withWidth(final int value) {
+            this.width = value;
+            return this;
+        }
+
+        /** Sets the height field. */
+        @JsonProperty("height")
+        public Builder withHeight(final int value) {
+            this.height = value;
+            return this;
+        }
+
+        /** Sets the format field. */
+        @JsonProperty("format")
+        public Builder withFormat(final String value) {
+            this.format = value;
+            return this;
+        }
+
+        /** Sets the exif field. */
+        @JsonProperty("exif")
+        public Builder withExif(final Map<String, String> value) {
+            this.exif = value;
+            return this;
+        }
+
+        /** Builds the ImageMetadata instance. */
+        public ImageMetadata build() {
+            return new ImageMetadata(
+                width,
+                height,
+                format,
+                exif
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/ImageMetadataType.java

@@ -0,0 +1,117 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Image element metadata.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ImageMetadataType.Builder.class)
+public record ImageMetadataType(
+    /**
+     * Image source (URL, data URI, or SVG content)
+     */
+    @JsonProperty("src") String src,
+    /**
+     * Alternative text from alt attribute
+     */
+    @Nullable @JsonProperty("alt") String alt,
+    /**
+     * Title attribute
+     */
+    @Nullable @JsonProperty("title") String title,
+    /**
+     * Image dimensions as (width, height) if available
+     */
+    @Nullable @JsonProperty("dimensions") List<Integer> dimensions,
+    /**
+     * Image type classification
+     */
+    @JsonProperty("image_type") ImageType imageType,
+    /**
+     * Additional attributes as key-value pairs
+     */
+    @JsonProperty("attributes") List<List<String>> attributes
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+private String src = "";
+private String alt = null;
+private String title = null;
+private List<Integer> dimensions = null;
+        @JsonProperty("image_type")
+private ImageType imageType = null;
+private List<List<String>> attributes = List.of();
+
+        /** Sets the src field. */
+        @JsonProperty("src")
+        public Builder withSrc(final String value) {
+            this.src = value;
+            return this;
+        }
+
+        /** Sets the alt field. */
+        @JsonProperty("alt")
+        public Builder withAlt(final @Nullable String value) {
+            this.alt = value;
+            return this;
+        }
+
+        /** Sets the title field. */
+        @JsonProperty("title")
+        public Builder withTitle(final @Nullable String value) {
+            this.title = value;
+            return this;
+        }
+
+        /** Sets the dimensions field. */
+        @JsonProperty("dimensions")
+        public Builder withDimensions(final @Nullable List<Integer> value) {
+            this.dimensions = value;
+            return this;
+        }
+
+        /** Sets the imageType field. */
+        @JsonProperty("image_type")
+        public Builder withImageType(final ImageType value) {
+            this.imageType = value;
+            return this;
+        }
+
+        /** Sets the attributes field. */
+        @JsonProperty("attributes")
+        public Builder withAttributes(final List<List<String>> value) {
+            this.attributes = value;
+            return this;
+        }
+
+        /** Builds the ImageMetadataType instance. */
+        public ImageMetadataType build() {
+            return new ImageMetadataType(
+                src,
+                alt,
+                title,
+                dimensions,
+                imageType,
+                attributes
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/ImagePreprocessingConfig.java

@@ -0,0 +1,142 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+
+/**
+ * Image preprocessing configuration for OCR.
+ *
+ * These settings control how images are preprocessed before OCR to improve
+ * text recognition quality. Different preprocessing strategies work better
+ * for different document types.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ImagePreprocessingConfig.Builder.class)
+public record ImagePreprocessingConfig(
+    /**
+     * Target DPI for the image (300 is standard, 600 for small text).
+     */
+    @JsonProperty("target_dpi") int targetDpi,
+    /**
+     * Auto-detect and correct image rotation.
+     */
+    @JsonProperty("auto_rotate") boolean autoRotate,
+    /**
+     * Correct skew (tilted images).
+     */
+    @JsonProperty("deskew") boolean deskew,
+    /**
+     * Remove noise from the image.
+     */
+    @JsonProperty("denoise") boolean denoise,
+    /**
+     * Enhance contrast for better text visibility.
+     */
+    @JsonProperty("contrast_enhance") boolean contrastEnhance,
+    /**
+     * Binarization method: "otsu", "sauvola", "adaptive".
+     */
+    @JsonProperty("binarization_method") String binarizationMethod,
+    /**
+     * Invert colors (white text on black → black on white).
+     */
+    @JsonProperty("invert_colors") boolean invertColors
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+    public ImagePreprocessingConfig{
+        if (targetDpi == 0) targetDpi = 300;
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("target_dpi")
+private int targetDpi = 0;
+        @JsonProperty("auto_rotate")
+private boolean autoRotate = true;
+private boolean deskew = true;
+private boolean denoise = false;
+        @JsonProperty("contrast_enhance")
+private boolean contrastEnhance = false;
+        @JsonProperty("binarization_method")
+private String binarizationMethod = "otsu";
+        @JsonProperty("invert_colors")
+private boolean invertColors = false;
+
+        /** Sets the targetDpi field. */
+        @JsonProperty("target_dpi")
+        public Builder withTargetDpi(final int value) {
+            this.targetDpi = value;
+            return this;
+        }
+
+        /** Sets the autoRotate field. */
+        @JsonProperty("auto_rotate")
+        public Builder withAutoRotate(final boolean value) {
+            this.autoRotate = value;
+            return this;
+        }
+
+        /** Sets the deskew field. */
+        @JsonProperty("deskew")
+        public Builder withDeskew(final boolean value) {
+            this.deskew = value;
+            return this;
+        }
+
+        /** Sets the denoise field. */
+        @JsonProperty("denoise")
+        public Builder withDenoise(final boolean value) {
+            this.denoise = value;
+            return this;
+        }
+
+        /** Sets the contrastEnhance field. */
+        @JsonProperty("contrast_enhance")
+        public Builder withContrastEnhance(final boolean value) {
+            this.contrastEnhance = value;
+            return this;
+        }
+
+        /** Sets the binarizationMethod field. */
+        @JsonProperty("binarization_method")
+        public Builder withBinarizationMethod(final String value) {
+            this.binarizationMethod = value;
+            return this;
+        }
+
+        /** Sets the invertColors field. */
+        @JsonProperty("invert_colors")
+        public Builder withInvertColors(final boolean value) {
+            this.invertColors = value;
+            return this;
+        }
+
+        /** Builds the ImagePreprocessingConfig instance. */
+        public ImagePreprocessingConfig build() {
+            return new ImagePreprocessingConfig(
+                targetDpi,
+                autoRotate,
+                deskew,
+                denoise,
+                contrastEnhance,
+                binarizationMethod,
+                invertColors
+            );
+        }
+    }
+    // CPD-ON
+    public static ImagePreprocessingConfig defaultInstance() {
+        throw new UnsupportedOperationException("defaultInstance is not yet bridged via JNI; use the Builder instead.");
+    }
+}

packages/java/dev/kreuzberg/ImagePreprocessingMetadata.java

@@ -0,0 +1,209 @@
+// This file is auto-generated by alef — DO NOT EDIT.
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+// To regenerate: alef generate
+// To verify freshness: alef verify --exit-code
+// Issues & docs: https://github.com/kreuzberg-dev/alef
+package dev.kreuzberg;
+
+import java.util.List;
+import com.fasterxml.jackson.annotation.JsonProperty;
+import com.fasterxml.jackson.annotation.JsonInclude;
+import com.fasterxml.jackson.databind.annotation.JsonDeserialize;
+import com.fasterxml.jackson.databind.annotation.JsonPOJOBuilder;
+import org.jspecify.annotations.Nullable;
+
+/**
+ * Image preprocessing metadata.
+ *
+ * Tracks the transformations applied to an image during OCR preprocessing,
+ * including DPI normalization, resizing, and resampling.
+ */
+@JsonInclude(JsonInclude.Include.NON_ABSENT)
+@JsonDeserialize(builder = ImagePreprocessingMetadata.Builder.class)
+public record ImagePreprocessingMetadata(
+    /**
+     * Original image dimensions (width, height) in pixels
+     */
+    @JsonProperty("original_dimensions") List<Long> originalDimensions,
+    /**
+     * Original image DPI (horizontal, vertical)
+     */
+    @JsonProperty("original_dpi") List<Double> originalDpi,
+    /**
+     * Target DPI from configuration
+     */
+    @JsonProperty("target_dpi") int targetDpi,
+    /**
+     * Scaling factor applied to the image
+     */
+    @JsonProperty("scale_factor") double scaleFactor,
+    /**
+     * Whether DPI was auto-adjusted based on content
+     */
+    @JsonProperty("auto_adjusted") boolean autoAdjusted,
+    /**
+     * Final DPI after processing
+     */
+    @JsonProperty("final_dpi") int finalDpi,
+    /**
+     * New dimensions after resizing (if resized)
+     */
+    @Nullable @JsonProperty("new_dimensions") List<Long> newDimensions,
+    /**
+     * Resampling algorithm used ("LANCZOS3", "CATMULLROM", etc.)
+     */
+    @JsonProperty("resample_method") String resampleMethod,
+    /**
+     * Whether dimensions were clamped to max_image_dimension
+     */
+    @JsonProperty("dimension_clamped") boolean dimensionClamped,
+    /**
+     * Calculated optimal DPI (if auto_adjust_dpi enabled)
+     */
+    @Nullable @JsonProperty("calculated_dpi") Integer calculatedDpi,
+    /**
+     * Whether resize was skipped (dimensions already optimal)
+     */
+    @JsonProperty("skipped_resize") boolean skippedResize,
+    /**
+     * Error message if resize failed
+     */
+    @Nullable @JsonProperty("resize_error") String resizeError
+) {
+    public static Builder builder() {
+        return new Builder();
+    }
+
+    // CPD-OFF
+    @JsonPOJOBuilder(withPrefix = "with", buildMethodName = "build")
+    public static final class Builder {
+
+        @JsonProperty("original_dimensions")
+private List<Long> originalDimensions = List.of();
+        @JsonProperty("original_dpi")
+private List<Double> originalDpi = List.of();
+        @JsonProperty("target_dpi")
+private int targetDpi = 0;
+        @JsonProperty("scale_factor")
+private double scaleFactor = 0.0;
+        @JsonProperty("auto_adjusted")
+private boolean autoAdjusted = false;
+        @JsonProperty("final_dpi")
+private int finalDpi = 0;
+        @JsonProperty("new_dimensions")
+private List<Long> newDimensions = null;
+        @JsonProperty("resample_method")
+private String resampleMethod = "";
+        @JsonProperty("dimension_clamped")
+private boolean dimensionClamped = false;
+        @JsonProperty("calculated_dpi")
+private Integer calculatedDpi = null;
+        @JsonProperty("skipped_resize")
+private boolean skippedResize = false;
+        @JsonProperty("resize_error")
+private String resizeError = null;
+
+        /** Sets the originalDimensions field. */
+        @JsonProperty("original_dimensions")
+        public Builder withOriginalDimensions(final List<Long> value) {
+            this.originalDimensions = value;
+            return this;
+        }
+
+        /** Sets the originalDpi field. */
+        @JsonProperty("original_dpi")
+        public Builder withOriginalDpi(final List<Double> value) {
+            this.originalDpi = value;
+            return this;
+        }
+
+        /** Sets the targetDpi field. */
+        @JsonProperty("target_dpi")
+        public Builder withTargetDpi(final int value) {
+            this.targetDpi = value;
+            return this;
+        }
+
+        /** Sets the scaleFactor field. */
+        @JsonProperty("scale_factor")
+        public Builder withScaleFactor(final double value) {
+            this.scaleFactor = value;
+            return this;
+        }
+
+        /** Sets the autoAdjusted field. */
+        @JsonProperty("auto_adjusted")
+        public Builder withAutoAdjusted(final boolean value) {
+            this.autoAdjusted = value;
+            return this;
+        }
+
+        /** Sets the finalDpi field. */
+        @JsonProperty("final_dpi")
+        public Builder withFinalDpi(final int value) {
+            this.finalDpi = value;
+            return this;
+        }
+
+        /** Sets the newDimensions field. */
+        @JsonProperty("new_dimensions")
+        public Builder withNewDimensions(final @Nullable List<Long> value) {
+            this.newDimensions = value;
+            return this;
+        }
+
+        /** Sets the resampleMethod field. */
+        @JsonProperty("resample_method")
+        public Builder withResampleMethod(final String value) {
+            this.resampleMethod = value;
+            return this;
+        }
+
+        /** Sets the dimensionClamped field. */
+        @JsonProperty("dimension_clamped")
+        public Builder withDimensionClamped(final boolean value) {
+            this.dimensionClamped = value;
+            return this;
+        }
+
+        /** Sets the calculatedDpi field. */
+        @JsonProperty("calculated_dpi")
+        public Builder withCalculatedDpi(final @Nullable int value) {
+            this.calculatedDpi = value;
+            return this;
+        }
+
+        /** Sets the skippedResize field. */
+        @JsonProperty("skipped_resize")
+        public Builder withSkippedResize(final boolean value) {
+            this.skippedResize = value;
+            return this;
+        }
+
+        /** Sets the resizeError field. */
+        @JsonProperty("resize_error")
+        public Builder withResizeError(final @Nullable String value) {
+            this.resizeError = value;
+            return this;
+        }
+
+        /** Builds the ImagePreprocessingMetadata instance. */
+        public ImagePreprocessingMetadata build() {
+            return new ImagePreprocessingMetadata(
+                originalDimensions,
+                originalDpi,
+                targetDpi,
+                scaleFactor,
+                autoAdjusted,
+                finalDpi,
+                newDimensions,
+                resampleMethod,
+                dimensionClamped,
+                calculatedDpi,
+                skippedResize,
+                resizeError
+            );
+        }
+    }
+    // CPD-ON
+}

packages/java/dev/kreuzberg/ImageProcessingException.java

@@ -0,0 +1,15 @@
+// DO NOT EDIT - auto-generated by alef
+// alef:hash:4e15143f4af1ae8bafbdb1506ef057da924484c66a19483966333558ad437e75
+package dev.kreuzberg;
+
+public class ImageProcessingException extends KreuzbergErrorException {
+    /** Creates a new ImageProcessingException with the given message. */
+    public ImageProcessingException(final String message) {
+        super(message);
+    }
+
+    /** Creates a new ImageProcessingException with the given message and cause. */
+    public ImageProcessingException(final String message, final Throwable cause) {
+        super(message, cause);
+    }
+}