A fast, helpful, and open-source document parser
LiteParse is a local tool that extracts structured text and layout information from PDF files, including coordinates for each piece of text and optional OCR for scanned pages. It is for applications that need to turn PDFs into Markdown, JSON, or plain text without relying on cloud services.
Latest release node-v2.18.0 — Node.js v2.18.0 · 4 Nov 2026
2 files for Codex, OpenCode and Claude Code: liteparse AGENTS.md, liteparse CLAUDE.md — 2,135 tokens loaded in every session.
▶ LiteParse: Local Document Parsing for AI Agents LlamaIndex · about run-llama/liteparse · on YouTube → AGENTS.md A 2,131 tok CLAUDE.md A 4 tok These files are run-llama/liteparse's own configuration — they tell Codex, OpenCode and Claude Code how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.