rag-anything

Python library for RAG over PDFs, Office files, images, tables and equations

Stars: 23.5k
License: MIT
Last commit: Oct 2026
Last release: Oct 2026
Language: Python

Missed the latest GitHub check; these facts may be older.

Overview

RAG-Anything is a Python framework built on LightRAG that parses mixed-content documents with MinerU, Docling or PaddleOCR. It runs separate processors for images, tables and equations, builds a multimodal knowledge graph, and answers queries with combined vector and graph retrieval. LLM, vision and embedding functions are supplied by the caller, with OpenAI-style examples.

Who it is for: Python developers building RAG over documents with figures, tables and formulas

Strengths

Handles images, tables and LaTeX equations with dedicated processors instead of text-only chunking
Selectable parsers: MinerU, Docling or PaddleOCR
Can insert pre-parsed content lists, skipping document parsing
VLM-enhanced query mode sends document images to a vision model

Weaknesses

Library only; the visible README shows no web UI, server or user accounts
Office formats require a separate LibreOffice install
Needs external LLM and vision model functions that you wire up yourself
RAM and GPU requirements are not stated; MinerU models download on first use

What it needs

Needs LightRAG, MinerU, LibreOffice (Office files), Pillow (image extra), ReportLab (text extra)
Models: OpenAI-compatible LLM, vision and embedding functions

Also in RAG and knowledge

See all 23

Also in RAG and knowledge
Rank	Project	Score
1	deeptutorSelf-hosted AI tutor with knowledge bases, guided learning, books and quizzes41.4k starsApache-2.0	79 out of 100
2	Open NotebookSelf-hosted NotebookLM alternative with podcasts and 20+ model providers40.2k starsMIT	75 out of 100
3	LightRAGGraph-based RAG framework with a REST API server and web UI40.1k starsMIT	75 out of 100
#4	RAGFlowRAG engine with deep document parsing, template chunking and agentic retrieval Live demo ↗ (opens in a new tab)92.2k starsApache-2.0	74 out of 100
#5	MaxKBEnterprise knowledge-base agent platform with RAG, workflows and MCP tools23k starsGPL-3.0	69 out of 100

How to choose

Match the ingestion formats you need (PDF with tables, Office, web, Confluence) before anything else.
Check which embedding models and vector stores are supported and whether they can run offline.
Look for citations in answers; without them, RAG output is hard to trust.

Health

Releases: Not checked
Top contributor: Not checked
Security advisories: Not checked
Package advisories: Not checked
OpenSSF Scorecard: Not checked

Not read for this project yet: each refresh reads the facts of a few projects. Bots and coding-agent accounts are left out of the commit counts. Not part of the score: how we check.

License in plain words

MIT Permissive

Use it at work: Yes
Change it and rebrand it: Yes
Sell it or host it for others: Yes

Your part: Keep the copyright and license notice in copies you ship.

A plain summary, not legal advice. Read the license on GitHub ↗ (opens in a new tab)

#13 of 23 in RAG and knowledge on Archestack, as of 15 Oct 2026.

In your README

The badge updates twice a day.

[![Archestack](https://archestack.com/badge/hkuds/rag-anything.svg)](https://archestack.com/best-of-ai/projects/rag-anything/)

How badges work

On your site or blog

A card with the rank, score and signals, in light or dark to match the reader.

<iframe src="https://archestack.com/embed/best-of-ai/rag-anything/" title="rag-anything on Archestack" width="400" height="270" style="border:0;max-width:100%" loading="lazy"></iframe>

Preview the card

Best of Open-Source AI on GitHub

Each review comes from the project's README and facts we check on GitHub. The fix form opens with rag-anything filled in.