MLX LM

MLX LM: LLM Inference and Fine-Tuning on Apple Silicon

The promise of running LLMs locally on a MacBook has been seductive but incomplete. Ollama and llama.cpp made it possible, but performance left …