# QuivrHQ/MegaParse

File Parser optimised for LLM Ingestion with no loss 🧠 Parse PDFs, Docx, PPTx in a format that is ideal for LLMs.

Repository: https://github.com/QuivrHQ/MegaParse
Canonical: https://ross.abutalabs.com/products/megaparse
Homepage: https://megaparse.com
Language: Python
License: Apache-2.0
License Family: permissive
Topics: docx, llm, parser, pdf, powerpoint
Last push: 2025-02-21T15:21:14+00:00

## Health v2 (maintenance only)
Score: 29/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 7, release rhythm 40, longevity 59
- inputs: {"age_days": 826, "days_push": 558, "days_rel": 565, "gap_med": 0, "n_releases_24m": 32}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7413, forks 417 (observed 2026-08-28T04:09:59.749717+00:00)

## What it is
MegaParse is a Python library that parses PDFs, Word, PowerPoint, Excel, CSV, and text documents into LLM-friendly formats with a focus on zero information loss, including tables, headers, footers, and images. It offers both standard parsing and a vision-based mode using multimodal models like GPT-4o and Claude, and can also be run as an API service.

## Use cases
- parse pdfs for llm ingestion
- convert docx to markdown for rag
- extract tables from pdf without losing data
- prepare powerpoint slides for a chatbot knowledge base
- build a document ingestion pipeline for a vector database
- run a self-hosted document parsing api

## When to choose
- you need high-fidelity document parsing for RAG or LLM pipelines
- your documents contain tables, headers, footers, or images that naive parsers drop
- you want a benchmarked parser with vision-model support
- you prefer an open-source, self-hostable alternative to paid parsing APIs

## When to avoid
- you only need plain-text extraction from simple text-based PDFs
- you cannot provide an OpenAI or Anthropic API key for the vision parser
- you need a fully offline solution without installing poppler and tesseract
- your environment runs Python below 3.11

## Facets
- artifact type: library
- maturity: active
- function: parser, pdf, ocr, rag, llm-inference
- domain: pdf, files, developer-tools
- platform: python, cross-platform, self-hosted
- tags: document-parsing, pdf-to-markdown, docx, pptx, llm-ingestion, data-preparation, natural-language-processing, retrieval-augmented-generation, docker

## Member repositories
- QuivrHQ/MegaParse (main) score 29

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:59.749717+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:37:46.969669+00:00, confidence not recorded.
  - readme: https://github.com/QuivrHQ/MegaParse (fetched 2026-08-28T04:09:59.749717+00:00, sha 52385b3a0adf)
  - homepage: https://megaparse.com (fetched 2026-08-29T08:33:29.828362+00:00, sha f1b197872b3d)
- Data as of 2026-08-30T08:39:29.467469+00:00.