Documents & research

Datalab

High-precision document parsing and extraction API powered by Chandra 2.5, Marker, and Surya that converts PDFs, scans, spreadsheets, and slides into Markdown, HTML, JSON, or structured fields with bounding boxes.

Free plan

Free

Free

20 USD monthly free allowance on a work email (10 USD on a personal email) against the full hosted API rate card

No credit cardCommercial useJSON / MDAPI available

Free plan & overview

Official docs

No card to start

Use the monthly allowance without a card; add a card to enable pay-as-you-go (invoiced every $15 or monthly)

Limits

Pricing lists 25 RPM / 25 concurrency (docs list 10 RPM / 5 concurrent), up to 200 MB and 7,000 pages per request, plus 1 custom processor creation/month

Best for

Engineering and data teams parsing contracts, invoices, financial statements, and complex tables at scale for RAG pipelines and automated document workflows.

How to get started

Sign up free at datalab.to to get $20/month in usage with a work email ($10/month with a personal email, no card required), then test in the Playground or call the REST API, Python SDK, or CLI.

Platforms & clients
REST API & Python SDK · Datalab CLI · Web Dashboard & Playground · APIOpen source · Self-hosting supported
Region & network
GlobalAvailable globally in USD; US processing by default, with EU data residency billed at 1.25x usage (+25%).
Payment & sign-in
Credit / debit card / InvoiceSign in: Email / SSO
Data & privacy
Enterprise: Not used for training;Free: Conditional;Team: Not used for training · Training opt-out availableCompliance: SOC 2 / HIPAA / GDPR
Rate limits
Pricing: Free 25 RPM / 25 concurrency, Team 400 RPM / 400 concurrency (Docs limits.md: Free 10 RPM / 5 concurrent, Team 200 RPM / 400 concurrent; 15,000 concurrent pages in flight per team)