LiteParse
Verifiedby LlamaIndex
Parses PDFs, Office files, and images locally into markdown, JSON with bounding boxes, or plain text for agents and RAG pipelines, from a CLI or Rust, Node.js, Python, and WASM libraries.
- Access
- CLIAPI
- Auth
- None
- Skills
- 5
Skills
Markdown Output
Renders documents to structured Markdown with headings, tables, lists, images, and links, ready to pass to an LLM or RAG pipeline.
Selective OCR
Runs bundled Tesseract OCR only where native text is missing, or sends pages to an HTTP OCR server such as EasyOCR or PaddleOCR.
Complexity Check
Checks cheaply whether a document needs OCR or heavier parsing, so a pipeline can route, reject, or estimate cost before a full parse.
Page Screenshots
Generates PNG screenshots of all or selected pages at a custom DPI so vision-capable agents can look at the rendered page itself.
Batch And Worker Parsing
Parses whole directories with lit batch-parse, or many files in Python or Node.js worker pools with hard per-parse timeouts.
Related Agents
MinerU
by OpenDataLab
Converts PDFs, Office documents, and scanned pages into LLM-ready Markdown or JSON, preserving tables, formulas, and…
MarkItDown
by Microsoft
Converts PDF, Office, HTML, image, audio, and ZIP files into structured Markdown for LLM pipelines, preserving…
Chandra
by Datalab
Converts PDFs and images into Markdown, HTML or JSON with layout preserved, handling tables, forms, handwriting and 90+…
PaddleOCR
by PaddlePaddle
Converts documents, receipts, and images into structured data with OCR, layout analysis, and table recognition, exposed…