run-llama/liteparse

Health v2 · maintenance only

Adoption not part of the score

12181 stars · 843 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

LiteParse is a fast, open-source document parser written in Rust that extracts spatial text with bounding boxes from PDFs, Office files, and images, entirely locally with no cloud dependencies. It offers a CLI and library APIs for Rust, Python, Node.js, and the browser (WASM), with built-in Tesseract OCR, pluggable OCR servers, Markdown/JSON/text output, and page screenshots.

Use cases

• parse pdfs to text locally
• extract text with bounding boxes from pdf
• convert pdf to markdown for rag pipelines
• ocr scanned documents offline
• parse docx xlsx pptx files
• generate pdf page screenshots for llm agents
• score pdf complexity before parsing
• run document parsing in the browser with wasm

When to choose

• you need fast, local PDF text extraction without cloud APIs or LLMs
• you need spatial layout info and bounding boxes for downstream processing
• you want Markdown output to feed LLMs or RAG pipelines
• you need a parser usable from Rust, Python, Node.js, or the browser

When to avoid

• you need to parse dense tables, multi-column layouts, charts, or handwriting accurately - a cloud parser like LlamaParse will do better
• you need cloud-scale or LLM-powered document extraction features
• you need browser parsing of complex scanned PDFs (WASM has limitations)

Facets

library · maturity active

parser ocr pdf image-processing cli sdk pdf files developer-tools artificial-intelligence rust python cli wasm browser cross-platform pdf-parsing text-extraction bounding-boxes tesseract rag-pipelines document-conversion wasm llamaindex natural-language-processing nodejs

Member repositories

For agents

markdown · JSON · MCP: product_card(name="run-llama/liteparse")