Full threadaccelbred·Looks cool!If it gets support for ollama or the llama-cpp server, I'll give it a go.View on HN