Native, Apple Silicon–only local LLM server. Similar to Ollama, but built on Apple's MLX for maximum performance on M‑series chips. SwiftUI app + SwiftNIO server with OpenAI‑compatible endpoints. - View it on GitHub