llm is a wrapper for interacting with locally run (or remote, via API) generative AIs from Simon Willison. Tastes great with gpt4all, which assists in downloading and running models locally.

Adam Keys @therealadam