Model docs
Every model OCRRouter routes to, with the upstream's raw response explained field by field. Field meanings come from each vendor's documentation, read on 2026-11-19.
- Mistral OCR
mistral/mistral-ocrPOST https://api.mistral.ai/v1/ocr25 credits / page28 fields - Gemini Flash
google/gemini-flashPOST https://generativelanguage.googleapis.com/v1beta/models/{model}:generateContent15 credits / page50 fields - GPT
openai/gptPOST https://api.openai.com/v1/responses11 credits / page57 fields - Claude
anthropic/claudePOST https://api.anthropic.com/v1/messages, anthropic-version 2023-06-0125 credits / page44 fields - Amazon Textract
aws/textractPOST https://api.openparser.dev/parse (multipart: request JSON + file)25 credits / page61 fields - Azure Document Intelligence
azure/document-intelligencePOST https://api.openparser.dev/parse (multipart: request JSON + file)60 credits / page75 fields - Google Document AI
google/document-aiPOST https://api.openparser.dev/parse (multipart: request JSON + file)10 credits / page70 fields - DeepSeek-OCR
deepseek/deepseek-ocrPOST https://aihubmix.com/v1/chat/completions1 credit / page17 fields - Jina OCR
jina/jina-ocrPOST https://aihubmix.com/v1/chat/completions1 credit / page17 fields - Qianfan OCR
baidu/qianfan-ocrPOST https://aihubmix.com/v1/chat/completions2 credits / page17 fields - PaddleOCR-VL
paddle/paddleocr-vlPOST https://api.openparser.dev/parse (multipart)6 credits / page37 fields - GLM-OCR
zai/glm-ocrPOST https://api.openparser.dev/parse (multipart)6 credits / page52 fields - Tesseract
tesseract/tesseracttesseract <image> stdout -l <langs>, run once per page imageFree13 fields