API docs / Text models
deepseek-ocr API
Converts images containing documents, PDFs, charts, and handwritten text into structured markdown while preserving formatting, tables, and layout. Handles complex multi-column documents, mathematical equations (converted to LaTeX), chemical formulas, scientific content, and multilingual text across 100 languages. Offers multiple task types including markdown conversion, raw text extraction, figure parsing, and content location by reference text. Uses visual sequence understanding to interpret document structure rather than simple character recognition, achieving high compression efficiency by processing documents as visual tokens. Supports various document types including academic papers, financial reports, receipts, invoices, newspapers, and handwritten notes.
- Model id
lucataco/deepseek-ocr- Provider
- lucataco
- Output
- Text
- Price
- From 2 credits per run, the same as on voor.ai. Options such as length or resolution can change it.
Request
POST https://voor.ai/v1/generations with "model": "lucataco/deepseek-ocr". Poll GET /v1/tasks/{id} or receive the signed webhook. See the overview for keys, idempotency, webhooks and errors.
curl https://voor.ai/v1/generations \
-H "Authorization: Bearer $VOOR_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: $(uuidgen)" \
-d '{
"model": "lucataco/deepseek-ocr",
"input": {
"reference_images": [
"https://example.com/input.png"
]
},
"callback_url": "https://example.com/webhooks/voor"
}'
# Then poll the task (or wait for the webhook):
curl https://voor.ai/v1/tasks/TASK_ID \
-H "Authorization: Bearer $VOOR_API_KEY"Input parameters
| Name | Type | Details |
|---|---|---|
reference_imagesrequired | array of URLs | Input image to perform OCR on (supports documents, charts, tables, etc.) Up to 1 files · Accepts image/* |
resolution | string | Default: One of: gundam (recommended), tiny, small, base, large |
reference_text | string | Reference text to locate in the image (only used for 'Locate Object by Reference' task). Example: 'the teacher', '20-10', 'a red car' Default: |
task_type | string | Default: One of: Convert to Markdown, Free OCR, Parse Figure, Locate Object by Reference |