# ai-chatbot

small ai-chatbot for LF6

## Tech Infra

We use Ollama to provide inference to a prefered model. \
There is a `docker-compose` in this repo.

## Setup

In the directory of the project execute: `python -m venv ./venv` \
Then execute in the same directory: `./venv/bin/pip install torch transformers gradio openai` \
You also have to install a model inside of ollama which is done by going in the docker container via `docker exec ollama -it /bin/bash`\
and then running for example `ollama pull llama3.2`