MLX

What MLX is

MLX is a free array framework for machine learning research, designed for Apple silicon and its unified memory architecture. It provides a NumPy-like API, neural network and optimizer packages, and transformations for automatic differentiation and graph optimization. Operations can run on supported CPU or GPU devices. The framework is intended for training and deploying models, with examples covering transformer training, LLaMA text generation, LoRA fine-tuning, Stable Diffusion image generation, and Whisper speech recognition. Language bindings include Python, Swift, C++, and C. MLX is listed for iOS, Linux, and macOS; installation guidance covers macOS through pip and Linux CUDA or CPU-only packages. It supports training and deployment, multiple deployment targets, GPU acceleration, and distributed training. Listed model formats are Safetensors and GGUF. The project was founded in 2023 and uses the MIT license. Documentation, quick-start guidance, examples, and contribution guidelines are linked. The MLX plan costs nothing.