Neura Market
Minima logo

Minima

Free

Chat with local documents, connect local environment to ChatGPT or Claude ![GitHub Repo stars](https://img.shields.io/github/stars/dmayboroda/minima?style=social)

FreeFree tier
Inputs: text, fileOutputs: text
Type
Open Source
Visit Website GitHub (40 stars)

About Minima

Minima is an open-source, on-premises RAG (Retrieval-Augmented Generation) system that runs in configurable Docker containers. It enables users to chat with their local documents securely by connecting either to a fully local LLM (via Ollama), a custom OpenAI-compatible API, or to external AI assistants like ChatGPT and Anthropic Claude via Model Context Protocol (MCP). The tool supports four deployment modes: isolated local setup, custom LLM, ChatGPT integration, and MCP usage for Claude. Users configure the system through environment variables and a simple run.sh script, making it easy to deploy containerized RAG without external dependencies if desired.

Key Features

On-premises RAG with configurable Docker containers
Four deployment modes: fully local (Ollama), custom LLM (OpenAI-compatible API), ChatGPT integration, and MCP usage for Claude
Supports integration with ChatGPT, Claude, and any OpenAI-compatible LLM
Simple setup via run.sh script or manual Docker Compose commands
Environment variables for flexible configuration (embedding, reranker, LLM models)
Works with local files; no external dependencies required in isolated mode
Open source (MIT license) with active development on GitHub

Pros & Cons

Pros
  • Fully open source and free to use
  • Multiple integration options (local, ChatGPT, Claude, custom LLM)
  • Runs entirely on-premises, ensuring data privacy and security
  • Containerized deployment simplifies setup and scaling
  • Active GitHub repository with clear documentation
  • Supports a variety of open-source and proprietary LLMs
Cons
  • Requires Docker knowledge and manual configuration
  • Setup can be complex for non-technical users
  • Limited to RAG-based document retrieval; not a general-purpose chatbot
  • Dependency on external services when using ChatGPT/Claude modes

Best For

Secure document Q&A on-premises without sending data to third-party APIsConnecting local document repositories to ChatGPT or Claude for queryingBuilding custom RAG applications with your own LLM serverPrivacy-sensitive enterprise deployments that require data sovereignty

FAQ

What is Minima?
Minima is an open-source, on-premises RAG system that runs in configurable containers, allowing users to chat with local documents using various LLM backends.
What modes does Minima support?
Minima supports four modes: fully local with Ollama, custom LLM via OpenAI-compatible API, ChatGPT integration, and MCP usage for Anthropic Claude.
How do I set up Minima?
You can use the run.sh script to select a mode, or manually run Docker Compose commands with a .env file. Detailed instructions are in the GitHub README.
Is Minima free?
Yes, Minima is completely free and open source under the MIT license.