Best open-source AI repos for RAG and chatting with your documents (2026)

What is the best open-source repo for RAG and chatting with your own documents?

The best open-source AI repos for RAG and chatting with your documents right now are Dify, Open WebUI and Graphify. Dify is the most complete builder (visual workflows plus a RAG pipeline); Open WebUI is the quickest way to chat with local documents next to a local model; Graphify is built for codebases and their docs.

Comparison

Repo	What it does	Stars	Start here
Dify langgenius/dify · Python	Dify is an open-source platform for building LLM apps.	159,456	Self-host Dify with Docker Compose Beginner, 20 minutes
Open WebUI open-webui/open-webui · Python	Open WebUI is a self-hosted AI platform with a chat interface for Ollama and OpenAI-compatible APIs that can run entirely offline.	156,670	Run a private ChatGPT-style app for your local models Beginner, 10 minutes
Graphify Graphify-Labs/graphify · Python	Graphify is a tool that turns a codebase, its docs, configs, and PDFs into a knowledge graph that an AI coding assistant can query instead of grepping files.	132,336	Map an unfamiliar codebase in one run Intermediate, 15 minutes
Utopia deeplethe/utopia · Rust	Utopia is an open-source, self-hosted enterprise knowledge base in Rust that stores facts with time awareness, so you can ask what was true at a given date and what the base knew then.	8,543	Find which contract terms applied on a given date Intermediate, 1 hour
Ciele MattiaIppoliti/ciele · TypeScript	Ciele is a self-hostable TypeScript platform for building embeddable AI chat assistants that answer from your own websites, files, and FAQs, with escalation to human help desks and conversation analytics.	327	Run a private AI assistant on your own machine Intermediate, 45 minutes

When to pick which

Dify langgenius/dify

Dify is an open-source platform for building LLM apps. It combines a visual workflow canvas, a RAG pipeline that ingests PDFs and slides, agents with their own sandbox, support for hundreds of models, a prompt IDE, observability and a backend-as-a-service API, so a team can go from prototype to production in one workspace. It runs in Dify Cloud or self-hosted with Docker Compose.

Best for: Teams and builders who want to ship chatbots, internal assistants and AI workflows without hand-writing the whole stack.

Needs: Docker and Docker Compose v2.24.0 or later for self-hosting; A machine with at least 2 CPU cores and 4 GiB RAM; API keys for the model providers you use

Start here: Self-host Dify with Docker Compose Beginner, 20 minutes

Clone the repo, then cd dify and cd docker
Run cp .env.example .env
Start it with docker compose up -d
Open http://localhost/install and complete initialization

All 3 use cases GitHub

Open WebUI open-webui/open-webui

Open WebUI is a self-hosted AI platform with a chat interface for Ollama and OpenAI-compatible APIs that can run entirely offline. It adds local RAG over your documents with hybrid search, web search from many providers, custom models and agents with tools and knowledge, voice and video calls, role-based access control for teams, and SQLite or PostgreSQL storage. It installs with pip, Docker or Kubernetes.

Best for: Individuals and teams who want a ChatGPT-style interface for local or cloud models on their own server.

Needs: Docker, or Python with pip; Ollama for local models, or an OpenAI-compatible API key; A GPU is optional (the :cuda image uses it)

Start here: Run a private ChatGPT-style app for your local models Beginner, 10 minutes

With Docker: docker run -d -p 3000:8080 --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:main
Or with Python: pip install open-webui, then open-webui serve
Open http://localhost:3000 and create the admin account
Pick an Ollama model, or add an OpenAI-compatible API in settings

All 3 use cases GitHub

Graphify Graphify-Labs/graphify

Graphify is a tool that turns a codebase, its docs, configs, and PDFs into a knowledge graph that an AI coding assistant can query instead of grepping files.

Best for: Developers and engineering leads who work in an unfamiliar or large codebase and use Claude Code, Cursor, Codex, Gemini CLI, or another supported assistant.

Needs: Python 3.10 or newer; uv (recommended) or pipx; A supported AI coding assistant, such as Claude Code, Cursor, Codex, or Gemini CLI; A local copy of the repository you want to map

Start here: Map an unfamiliar codebase in one run Intermediate, 15 minutes, works with claude-code, cursor, codex, gemini-cli

Run uv tool install graphifyy (or pipx install graphifyy) to install the CLI.
Run graphify install once so it registers the skill with your assistant.
Open your AI assistant in the repository and type /graphify .
Open graphify-out/graph.html in a browser and read graphify-out/GRAPH_REPORT.md.

All 4 use cases GitHub

Utopia deeplethe/utopia

Utopia is an open-source, self-hosted enterprise knowledge base in Rust that stores facts with time awareness, so you can ask what was true at a given date and what the base knew then.

Best for: Engineers and knowledge or operations staff at a company who need dated, evidence-backed answers from their own documents and can run Docker and configure chat and embedding model endpoints.

Needs: Docker installed (local development also needs Rust 1.85+, Node 20+, pnpm); A chat model endpoint and an embedding model endpoint, for example any OpenAI-compatible endpoint; Business documents to ingest, such as PDF or Word contracts and amendments

Start here: Find which contract terms applied on a given date Intermediate, 1 hour

Run git clone https://github.com/deeplethe/utopia.git and then cd utopia.
Run docker compose --profile app up -d and open http://localhost:1516 to register; the first account becomes the administrator.
Under Administration, go to Models and configure the chat and embedding endpoints before extracting business documents.
Create a knowledge base, upload the agreement and its amendments as PDF or Word, and let extraction finish.

All 4 use cases GitHub

Ciele MattiaIppoliti/ciele

Ciele is a self-hostable TypeScript platform for building embeddable AI chat assistants that answer from your own websites, files, and FAQs, with escalation to human help desks and conversation analytics.

Best for: Developers, support leads, and operations teams who want a private AI assistant grounded in their own content and need to run it themselves or hand the setup to an agent.

Needs: A shell on a machine where ./deploy/bootstrap.sh can run; Git to clone the repository; Docker Desktop, if you use the Ciele Desktop app on macOS; Optional: an OpenAI chat and embeddings compatible server, such as a local model runner, if you do not want to use a hosted AI vendor

Start here: Run a private AI assistant on your own machine Intermediate, 45 minutes

Clone the Ciele repository from GitHub (the README clone line is truncated here, so use the repo URL shown on the GitHub page), then run cd ciele.
Run ./deploy/bootstrap.sh to generate secrets, start the database, apply migrations, and boot the app.
Open the app address shown in the output, sign up, and keep the account you create as the organization owner.
Optional: run ./deploy/bootstrap.sh --seed to start from populated demo content.

All 4 use cases GitHub

Use cases for RAG and chatting with your documents

Self-host Dify with Docker Compose Dify
Build a chatbot over your PDFs and slides Dify
Give a Dify agent tools from MCP servers and the marketplace Dify
Run a private ChatGPT-style app for your local models Open WebUI
Chat with your own documents using local RAG Open WebUI
Share one AI workspace with your team, with roles and groups Open WebUI
Map an unfamiliar codebase in one run Graphify
Answer architecture questions from the graph Graphify
Trace how two components connect before a change Graphify
Find which contract terms applied on a given date Utopia
Show who held a role in a past quarter and what was known then Utopia
Connect a coding agent to a knowledge base over MCP Utopia
Run a private AI assistant on your own machine Ciele
Build a support assistant from your website and FAQs Ciele
Audit assistant answers from conversation transcripts Ciele

Questions

What is the best open-source repo for RAG and chatting with your own documents?

Which repo for RAG and chatting with your documents is easiest to start with?

Dify: its starter use case "Self-host Dify with Docker Compose" is rated Beginner and takes about 20 minutes.

How do I get started with Dify?

Clone the repo, then cd dify and cd docker; Run cp .env.example .env; Start it with docker compose up -d. Full steps and a copy-paste agent prompt: https://dedupedai.com/use-cases/langgenius-dify/self-host-dify-with-docker-compose/

Are these RAG and chatting with your documents repos free?

They are open-source projects on GitHub (check each license; some, like fair-code projects, limit commercial use). Some also sell a hosted version or need a paid model API key; the requirements for each repo are listed on this page.

How is this list of RAG and chatting with your documents repos chosen?

Deduped tracks the hottest AI repos on GitHub and writes step-by-step use cases from each README. For this page we picked the repos that do this job, ordered by fit and then GitHub stars. Repos we only inferred from a short README, or that we do not index, are left out. The page is rebuilt with the site; stars are as of the last build in 2026.

What is the best open-source repo for RAG and chatting with your own documents?

Comparison

When to pick which

Dify langgenius/dify

Open WebUI open-webui/open-webui

Graphify Graphify-Labs/graphify

Utopia deeplethe/utopia

Ciele MattiaIppoliti/ciele

Use cases for RAG and chatting with your documents

Questions

More answers