LiteParse
LlamaIndex's open-source document parser — fully local structured-text extraction
LlamaIndex's open-source document parser that extracts structured text from PDF/Office files fully locally; the Rust rewrite made small docs 100× faster.
Translate documents, parse PDFs, and extract text with OCR.
Key features
- Fast lightweight PDF parsing: about 2-5ms per page using PDFium
- Spatial text parsing with precise bounding boxes
- Flexible OCR: built-in Tesseract or pluggable HTTP OCR servers
- Outputs Markdown, JSON and text; works in Rust, Python, Node, browser
Good for
- Developers needing local PDF parsing without LLMs or cloud
- RAG pipelines feeding structured Markdown to models
Stack & tags
RustPythonNode.jsTypeScriptpdf-parsingfastlocal-firstocrbounding-boxesmarkdownrust
Similar open-source projects
微软 MarkItDown
★ 187,737Microsoft's open-source Python tool — convert PDF/Word/PPT/audio/YouTube to clean Markdown
DeepSeek-OCR
★ 23,926DeepSeek's novel OCR document model — "contextual optical compression"