minima
Minima is an open-source RAG system that runs in containers on your own machine or cloud. It indexes a local folder and lets you ask questions about the documents, with four ways to connect the answering model.
In the fully local mode, everything (the LLM, the reranker and the embedding model) runs through Ollama in your containers, with no external services. A custom-LLM mode keeps the indexer local and uses any OpenAI-compatible endpoint such as vLLM, an Ollama server or TGI. A Custom GPT mode lets you query your documents from the ChatGPT app, and an MCP mode lets the Claude desktop app use Minima's MCP server while the indexer stays on your machine.
The easiest start is `./run.sh`, which offers the four options in a menu; alternatively you fill in `.env` (local files path, embedding model and size, and the model settings for your mode) and run the matching Docker Compose file. GitHub does not detect a licence, and our data shows the last push on 2026-01-22.
Features
Why choose it
Trade-offs
Maintenance check
- Last commit
- 2026-01-22
- Commits, last 90 days
- 0
- Releases, last 90 days
- 0
- Latest release
- beta (2025-01-22)
- Open issues
- 16
- Recent issues with at least one comment
- 5/17
Oldest week on the left; the latest four weeks are orange. Source: GitHub commit activity.
Status rule: archived on GitHub = Archived; last commit within 90 days = Active; 91 to 365 days = Slowing; older = Stale. All figures come from the GitHub API on the date shown.
Most discussed open issues
- #36 Chat gives no answer4 comments · 2025-02-21
- #11 Problem in vector DB.2 comments · 2024-12-04
- #57 Minima (Local RAG) has a Nerq Trust Score of 69.8/100 — verified badge available0 comments · 2026-05-12
- #56 Your project is listed on Spark — claim your listing0 comments · 2026-03-31
- #49 llm ERROR: Exception in ASGI application after startup0 comments · 2025-06-24
Issue titles are quoted from GitHub. Open the link to check whether a problem still applies to the version you use.
Compatibility
Quick start
Use cases
Alternatives
Comments
No comments yet. Be the first!