Overview
PaddleOCR-VL is a compact (0.9B) vision-language document parser from Baidu's PaddlePaddle team. A layout model finds blocks, then the VL model reads each one.
OCRRouter calls version 1.6 through OpenParser and returns page Markdown plus layout blocks with pixel boxes and detection scores.
Price and limits
- Model ID
- paddle/paddleocr-vl
- Price
- 6 credits / page ($6 per 1,000 pages)
- Max pages per request
- 500
- Inputs
- PDF; PNG, JPEG and other images OpenParser accepts
- Languages
- 109 languages
- Positions (bbox)
- Yes, when the upstream returns them
- License
- Apache-2.0
- Up to 500 pages per job in OCRRouter
- Confidence is the layout detection score, not a text recognition score
- Model can be changed server-side with OPENPARSER_PADDLE_MODEL
You pay only for pages that succeed. A failed attempt that falls back to another model is not charged. See pricing.
How OCRRouter calls it
- Gateway
- OpenParser
- Endpoint
- POST https://api.openparser.dev/parse (multipart)
- Upstream model
- ocr_model "paddleocr-vl-1.6"
- Request
- request part {"ocr_model", "output_format": "openparser@1"} plus the original file; Idempotency-Key header per call
- Long documents
- HTTP 202 with a job; OCRRouter polls /jobs/{id} with backoff from 1 to 5 seconds
Options
What OCRRouter sets on the upstream call. You choose the model, format and pages in your request; vendor-specific options are not passed through.
- ocr_model
- "paddleocr-vl-1.6"
- output_format
- "openparser@1"
- File
- Sent as uploaded, no rendering
- Client options
- None. Callers cannot pass model options.
Raw response fields
Every field the upstream documents, with its type and meaning taken from the vendor's own reference. Fields marked "read by OCRRouter" are the ones our connector uses; the rest are listed so you know what the engine produces.
What OCRRouter receives (OpenParser openparser@1)
OpenParser runs PaddleOCR-VL and converts its result into the openparser@1 document graph: pages with reading order, and elements with locations and scores. Field meanings follow the openparser@1 schema reference.
Source: openparser@1 schema, retrieved 2026-11-19
output_formatstring- Schema version, always "openparser@1".Example:
"openparser@1" document_idstring- Stable identifier for the parsed document.
provenance.provider / provenance.modelstring- Provider and model that produced the result.Example:
"baidu" / "paddleocr-vl-1.6" textstring, read by OCRRouter- Plain text in reading order. Spans index into it.
markdownstring, read by OCRRouter- Best-effort Markdown rendering of the whole document.
pages[].numberinteger, read by OCRRouter- One-based page number, contiguous 1 to N.Example:
1 pages[].width / heightnumber, read by OCRRouter- Size of the page coordinate space.Example:
800 / 300 pages[].unitstring, read by OCRRouter- Unit of the coordinate space: pixel, point, inch or normalized.Example:
"pixel" pages[].element_idsstring[], read by OCRRouter- IDs of elements on the page.
pages[].reading_orderstring[], read by OCRRouter- Ordered top-level element IDs for the page.
elements[].idstring, read by OCRRouter- Element ID, unique in the document.
elements[].kindstring, read by OCRRouter- Element type discriminator: text, table, figure, formula and others.Example:
"text" elements[].rolestring, read by OCRRouter- For text elements: heading, paragraph, line, word, page_header, page_footer and more.Example:
"heading" elements[].textstring, read by OCRRouter- Text of the element.Example:
"Total due: $42.50 by December 1." elements[].locations[].page_numberinteger, read by OCRRouter- Page the location is on.Example:
1 elements[].locations[].bboxobject, read by OCRRouter- Axis-aligned box {left, top, right, bottom} in page coordinates; right and bottom are exclusive.Example:
{"left":37,"top":107,"right":401,"bottom":135} elements[].locations[].polygonPoint[]- Tighter outline when the provider supplies one.
elements[].confidence.scorenumber, read by OCRRouter- Score normalized to 0 to 1.Example:
0.8717 elements[].confidence.scopestring- What the score measures: detection, recognition, classification, geometry, answer or quality.Example:
"detection" elements[].confidence.calibratedboolean- False means scores from different providers or scopes are not comparable probabilities.Example:
false elements[].confidence.source_value / source_scalenumber / string- Original provider value and its scale.Example:
0.8717 / "zero_to_one" elements[].source.native_id / native_type / native_labelstring- The engine's own ID, type and label for the block. Not in the published schema page; present in responses.Example:
"1" / "layout_block" / "text" elements[].row_count / column_count / cells[]number / array, read by OCRRouter- For table elements: grid size and cells with row_index, column_index, text.
relations[], text_annotations[], assets[]array- Links between elements, span metadata, and page or figure images.Example:
[]
Real response captured 2026-11-19 from POST /parse on an 800x300 PNG with three lines of text. Trimmed to the first two elements; nothing else edited.
{"output_format": "openparser@1","document_id": "3fd622b1876c999f4f0c3e48a415c085","provenance": {"provider": "baidu","model": "paddleocr-vl-1.6","operation": "layout_parsing"},"text": "## Invoice 2026\n\nTotal due: $42.50 by December …","markdown": "## ## Invoice 2026\n\nTotal due: $42.50 by Decemb…","pages": [{"number": 1,"source_page_number": 1,"width": 800,"height": 300,"unit": "pixel","rotation_degrees": 0,"languages": [],"element_ids": ["3fd622b1876c999f4f0c3e48a415c085-paddle-1-0","3fd622b1876c999f4f0c3e48a415c085-paddle-1-1"],"reading_order": ["3fd622b1876c999f4f0c3e48a415c085-paddle-1-0","3fd622b1876c999f4f0c3e48a415c085-paddle-1-1"]}],"elements": [{"id": "3fd622b1876c999f4f0c3e48a415c085-paddle-1-0","locations": [{"page_number": 1,"bbox": { "left": 38, "top": 28, "right": 247, "bottom": 64 },"polygon": [{ "x": 38, "y": 28 },{ "x": 247, "y": 28 },{ "x": 247, "y": 64 },{ "x": 38, "y": 64 }]}],"confidence": {"score": 0.7849714159965515,"scope": "detection","calibrated": false,"source_value": 0.7849714159965515,"source_scale": "zero_to_one"},"source": {"native_id": "0","native_type": "layout_block","native_label": "paragraph_title"},"kind": "text","role": "heading","text": "## Invoice 2026","spans": [],"languages": []},{"id": "3fd622b1876c999f4f0c3e48a415c085-paddle-1-1","locations": [{"page_number": 1,"bbox": { "left": 37, "top": 107, "right": 401, "bottom": 135 },"polygon": [{ "x": 37, "y": 107.00000762939453 },{ "x": 400.1377258300781, "y": 109.00076293945312 },{ "x": 399.9891052246094, "y": 135.9779052734375 },{ "x": 36.85137939453125, "y": 133.97715759277344 }]}],"confidence": {"score": 0.871687114238739,"scope": "detection","calibrated": false,"source_value": 0.871687114238739,"source_scale": "zero_to_one"},"source": {"native_id": "1","native_type": "layout_block","native_label": "text"},"kind": "text","role": "paragraph","text": "Total due: $42.50 by December 1.","spans": [],"languages": []}],"text_annotations": [],"relations": [],"assets": []}PaddleOCR-VL native output (pipeline result JSON)
What the PaddleOCR-VL pipeline itself returns per page, from save_to_json(). OpenParser turns parsing_res_list into elements and layout_det_res scores into confidence. OCRRouter never sees this shape directly.
Source: PaddleOCR-VL pipeline docs, retrieved 2026-11-19
input_pathstring- Path of the input image or PDF.
page_indexinteger or null- Current page for a PDF, else None.
page_countinteger or null- Total pages for a PDF, else None.
width / heightinteger- Size of the original input image.
model_settingsobject- Run settings such as use_doc_preprocessor, use_layout_detection, use_chart_recognition, format_block_content.
markdown_ignore_labelsstring[]- Labels left out of Markdown, by default number, footnote, header, header_image, footer, footer_image, aside_text.
parsing_res_list[]array- Parsed blocks; list order is the reading order.
parsing_res_list[].block_labelstring- Region label such as text, table, paragraph_title.
parsing_res_list[].block_contentstring- Recognized content of the region.
parsing_res_list[].block_bboxnumber[4]- Box of the layout region.
parsing_res_list[].block_id / block_orderinteger- Region index, and reading-order position (None when not sorted).
layout_det_res.boxes[]array- Raw layout detections, each with cls_id, label, score and coordinate.
layout_det_res.boxes[].scorenumber- Detection confidence of the box.
Built from the documented field list, with values from the real OpenParser response above. cls_id is illustrative. Not a captured PaddleOCR response.
{"input_path": "paddleocr_vl_demo.png","page_index": null,"page_count": null,"width": 800,"height": 300,"model_settings": {"use_doc_preprocessor": false,"use_layout_detection": true,"use_chart_recognition": false,"format_block_content": false},"parsing_res_list": [{"block_label": "paragraph_title","block_content": "Invoice 2026","block_bbox": [38, 28, 247, 64],"block_id": 0,"block_order": 1}],"layout_det_res": {"input_path": null,"page_index": null,"boxes": [{"cls_id": 17,"label": "paragraph_title","score": 0.785,"coordinate": [38, 28, 247, 64]}]}}How it maps to OCRRouter output
- pages[] sorted by number
- One OCRRouter page each
- pages[].reading_order (else element_ids)
- Order of layout blocks and Markdown on the page
- elements[].kind + role
- Block type: table, formula, figure/image/picture -> image, heading/title -> heading, else paragraph
- elements[].text (or markdown)
- Block text. Headings are normalized to one run of #, level 1 to 6
- Table cells
- Rebuilt as a GitHub-flavoured Markdown table; first row is the header
- locations[].bbox
- Block bbox [left, top, right, bottom], using the location on the same page
- pages[].width, height, unit
- Page layout width, height and unit (pixel here)
- confidence.score
- Block confidence (0 to 1); null when absent
- Page Markdown
- Block texts joined with blank lines
- Document markdown
- Fallback only when no element has text: all of it goes on page 1
The unified output is described in Response format.
Example request
curl https://api.ocrrouter.ai/v1/ocr \
-H "Authorization: Bearer $OCRROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "paddle/paddleocr-vl",
"file_url": "https://example.com/invoice.pdf",
"format": "json"
}'Sources
Vendor documentation read on 2026-11-19.