The unified API
for document parsing.
Any model, your docs.
Turn PDFs and images into markdown. Choose your model, send a document, get parsed pages back.
Documents go in,
structured markdown comes out.
Headings, paragraphs, tables, figures.
Get markdown for each page, ready to use in your application or pipeline.
River survey
The river bends east below the stone bridge. A heron stands in the shallows.
Along the bank
Willow, alder, and meadow grass line the water’s edge.
# River survey
The river bends east below the stone bridge.
A heron stands in the shallows.
## Along the bank
Willow, alder, and meadow grass line
the water’s edge.Illustrative input and output. Results vary by model and document.
Your model, your choice.
Choose a model for your document and budget. Every model is versioned with a parsing recipe and benchmarks from ParseBench.
Every page, as fast as possible.
Pages run as individual model calls as capacity becomes available. Each gets its own markdown, status, and charge.
Pick up where you left off.
Retry the pages that failed. Failed pages are free and page selection lets you skip the pages you already have.
A few lines.
A whole document.
Pick a model, point to your file, and choose your pages. We take care of the parsing recipe.
Explore the APIcurl https://www.opendocrouter.ai/v1/parse \
-H "Authorization: Bearer $OPEN_DOC_ROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "google/gemini-3.8-flash-low",
"document": {
"url": "https://arxiv.org/pdf/1706.03762"
},
"pages": "1-3"
}'Start without the commitment.
New accounts start with $5 of free credit, and top-ups start at $10. Frontier models cost exactly what their providers charge, and only the pages that succeed are charged.