Hey y'all! Since the Stanford Generative Agents source code is out now does anyone want to work together on converting it to use a locally-hosted LLM?