1. Change the example values in the provided `env.example` and rename it to `.env`.
2. Make sure ollama is running on the host and reachable at `localhost:11434` and the model that you specified in the `.env` file is already downloaded and ready to go. (Can be checked with e.g. `ollama run wizard-vicuna-uncensored`)
3. Run:
- With docker compose (might need sudo infront if you haven't setup rootless):
-`$ make up`
- Without docker:
-`$ make build && make run`
4. Join the Twitch channels you chose and type `()gpt <cool query>` and hopefully get a response.
