Repolode
Submit
jonfairbanks

jonfairbanks/local-rag

Offline RAG app that ingests local files, GitHub repos and websites and answers questions with local Ollama models, keeping data on your network.

View on GitHub
Stars
765
Forks
95
Open issues
10
Last commit
5 days ago

What is it?

Local RAG is an open-source retrieval augmented generation app that lets you chat with your own documents without sending content to third-party services. You point it at local files, GitHub repositories or websites, and it indexes that content so that a language model can answer questions grounded in it.

Chat uses local Ollama models, and embeddings can come from Ollama or from local Hugging Face models. Responses stream back through LlamaIndex. Settings are persisted in the browser, and chat history can be exported.

The project includes guardrails on uploads, URLs, repository sizes and ingestion, and its documentation covers setup, usage, the RAG pipeline and troubleshooting.

Key features

  • 01Chat with local Ollama models
  • 02Embeddings from Ollama or local Hugging Face models
  • 03Ingests local files, GitHub repositories and websites
  • 04Streams RAG responses through LlamaIndex
  • 05Exports chat history and persists settings in the browser

Use cases

  • Querying private documents without sending them to a cloud API
  • Asking questions about a codebase or GitHub repository offline
  • Building a local knowledge base from web pages and files

Things to know

  • Licensed under GPL-3.0, a copyleft license that requires derivative distributions to remain open source.
  • Chat models require a separately running Ollama instance with local models.

In collections

Private AI knowledge bases and RAGSelf-hosted appsRun LLMs locallyPrivacy & local-first tools

Topics

#rag#llm#ollama#retrieval-augmented-generation#embeddings#offline#self-hosted#large-language-models

FAQ

What is local-rag?
Offline RAG app that ingests local files, GitHub repos and websites and answers questions with local Ollama models, keeping data on your network. Local RAG is an open-source retrieval augmented generation app that lets you chat with your own documents without sending content to third-party services. You point it at local files, GitHub repositories or websites, and it indexes that content so that a language model can answer questions grounded in it.
Is local-rag free to use? What license does it have?
local-rag is open source under the GPL-3.0 license. Check the license terms before using it in commercial products.
Is local-rag still maintained?
Actively maintained. The last commit to local-rag was on Nov 2, 2026.
What can I use local-rag for?
Querying private documents without sending them to a cloud API Asking questions about a codebase or GitHub repository offline Building a local knowledge base from web pages and files
Where can I get local-rag?
The source code and installation instructions are on GitHub: https://github.com/jonfairbanks/local-rag

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects