Minima FAQs

Question 1

What is Minima and what does it do?

Accepted Answer

Minima is an open-source tool that allows you to run Retrieval Augmented Generation (RAG) pipelines on your own infrastructure using Docker containers. This means you can query your local documents using AI models without sending your data to external services, ensuring data privacy and control.

Question 2

Can I use Minima with ChatGPT or Anthropic Claude?

Accepted Answer

Yes! Minima supports integration with both ChatGPT and Anthropic Claude. You can use your local documents as context for ChatGPT via Custom GPTs or query them using the Anthropic Claude desktop app.

Question 3

What kind of documents can I use with Minima?

Accepted Answer

Minima supports a variety of document types, including `.pdf`, `.xls`, `.docx`, `.txt`, `.md`, and `.csv`. The indexing process is recursive, meaning all documents within subfolders of your specified root folder will also be indexed.

Question 4

What are the main features of Minima?

Accepted Answer

Minima offers several key features, including fully local RAG operations, integration with ChatGPT and Claude, Docker container deployment, support for multiple embedding models and LLMs (via Ollama), and a local Electron app for a chat UI with a fully local installation.

Question 5

How do I install and run Minima?

Accepted Answer

Minima can be installed and run using Docker Compose. You'll need to create a `.env` file with the required environment variables (like `LOCAL_FILES_PATH`, `EMBEDDING_MODEL_ID`, etc.) and then use Docker Compose to start the containers. There are specific `docker-compose.yml` files for fully local setups, ChatGPT integration, and Claude integration.

Minima

Minima

Key Features

Use Cases

Key Features

Use Cases