# LiteParse

> Apache-2.0 local PDF/doc parser — PDFium + OCR, no cloud required

**Category:** AI Agents  
**Pricing:** open-source  
**URL:** https://developers.llamaindex.ai/liteparse/  
**GitHub:** https://github.com/run-llama/liteparse  
**Added:** 2026-09-30

---

## The Hook

LiteParse is LlamaIndex’s fast local document parser — spatial text, bounding boxes, optional OCR, Markdown/JSON out — so agents ingest PDFs without a parse SaaS.

## What's Good

Apache-2.0 (run-llama/liteparse). Rust core with Python, Node, and WASM bindings. ~2–5ms/page PDFium path, bundled Tesseract or HTTP OCR, complexity detection before heavy parse, page screenshots for multimodal agents, worker-pool mode with per-parse timeouts. Docs at developers.llamaindex.ai/liteparse.

## The Catch

Vs LlamaIndex (framework) and cloud LlamaParse: LiteParse is the local OSS parser binary/bindings; LlamaParse cloud is the upsell for dense tables/scans/handwriting. Hard docs still want LlamaParse or another heavy parser — do not pretend local PDFium equals production OCR SaaS.

## Verdict

Apache local parse when agents need fast PDF→Markdown on-box — not LlamaParse cloud; escalate hard scans deliberately.