A simple Python and Docker-based app for iterative data analysis using an LLM.
Install Docker and WSL.
Clone this repository and then navigate in the terminal to the directory where it is saved. Then run
docker-compose up --build
When you open the application, make sure to use the correct API key for the model you select.
If you don't want to keep pasting your API key for each analysis, you can set API keys as environment variables and just select the model at runtime. There are currently 4 supported LLMs from 3 different providers:
- GPT-4o and o1 from OpenAI
- Claude 3.7 Sonnet from Anthropic
- Gemini 2.5 Pro from Google
Each provider will give you an API key to use their models. To set your Gemini API key, run this:
set API_KEY_GEM=YOUR-GEMINI-API-KEY
To set your API key for Claude, run this:
set API_KEY_ANT=YOUR-ANTHROPIC-API-KEY
And to set your API key for either of the OpenAI models, run this:
set API_KEY_OAI=YOUR-OPENAI-API-KEY
You can set all of these at the beginning and then use a different model each time you restart, without ever having to paste your API key into the GUI.
The application will run in localhost:5000.