26 lines
631 B
JSON
26 lines
631 B
JSON
|
{
|
||
|
|
"document": "../../../test_documents/images/english_and_korean.png",
|
||
|
|
"file_type": "png",
|
||
|
|
"file_size": 305401,
|
||
|
|
"expected_frameworks": [
|
||
|
|
"kreuzberg",
|
||
|
|
"docling",
|
||
|
|
"markitdown",
|
||
|
|
"mineru",
|
||
|
|
"pymupdf4llm",
|
||
|
|
"tika",
|
||
|
|
"unstructured"
|
||
|
|
],
|
||
|
|
"metadata": {
|
||
|
|
"description": "Scanned multilingual document with English and Korean text - 305KB PNG",
|
||
|
|
"category": "image-scanned",
|
||
|
|
"size_class": "medium",
|
||
|
|
"requires_ocr": true,
|
||
|
|
"languages": ["en", "ko"]
|
||
|
|
},
|
||
|
|
"ground_truth": {
|
||
|
|
"text_file": "../../../test_documents/ground_truth/png/image_scanned.md",
|
||
|
|
"source": "mistral-pixtral"
|
||
|
|
}
|
||
|
|
}
|