anydoc is Firecrawl's open-source document-parsing engine for RAG pipelines, launched on August 4, 2026. Firecrawl built it to replace heavier parser stacks that stall on mixed file types, and it turns messy PDFs, Word documents, slide decks, and images into clean structured Markdown in under five milliseconds per document, already powering Firecrawl's /parse endpoint.
Core Features
- Rust-backed engine that parses PDFs, Word files, presentations, and images to Markdown in under 5ms each.
- 100% open source, so teams can self-host it inside their own data pipeline.
- Clean, structured output tuned for retrieval-augmented generation rather than raw text dumps.
- Built to sit behind a single /parse API call, matching Firecrawl's hosted extraction flow.
Use Cases / Best For
- RAG developers who need reliable chunking from mixed document formats before embedding.
- Data-pipeline and knowledge-management engineers standardizing ingestion across file types.
- Teams that want fast, local parsing without sending documents to a third party.
Pros & Cons
- Pro: Sub-5ms, self-hostable parsing that keeps sensitive documents inside your own pipeline.
- Con: Running it yourself means you own uptime and updates unless you pay for Firecrawl's hosted /parse.
Pricing
anydoc itself is free and open source to self-host. It backs Firecrawl's /parse endpoint, which sits behind Firecrawl's free tier and paid plans for teams that prefer the hosted service over running the engine themselves.
Our Take: Best for RAG builders who want sub-5ms, self-hostable parsing without vendor lock-in; the trade-off is that you run and maintain it yourself if you skip Firecrawl's hosted tier. Find more developer tooling in our AI Programming development category.
