DocsRouter is an early-access OCR API for extracting text and tables from images and multi-page documents.
A curated model catalog
Instead of exposing every upstream model, DocsRouter maintains an allowlist of six OCR-capable vision models. Use GET /v1/models to discover the current IDs and pricing metadata.
Explicit or policy-based selection
Pass a supported model ID to /v1/ocr, or choose quality, accuracy, speed, cost, or balanced. These policies map to fixed documented models. They do not compare models or retry low-confidence output.
Transparent billing
Vision-model requests use the provider cost plus a 5% DocsRouter platform fee. Responses include the provider cost, platform fee, and total cost. Mistral OCR's dedicated parser has a direct provider rate of $2 per 1,000 pages before the platform fee.
Early-access boundaries
DocsRouter does not currently provide a result cache, adaptive fallback, benchmark leaderboard, or accuracy and latency guarantees. Test the supported models against your own documents and validate output before taking consequential action.
Make your first request or review the Native OCR API.