# Awesome MLX

A curated list of articles, models, libraries, and tools built around Apple's MLX machine learning framework for Apple Silicon.

Awesome MLX is a curated list of resources for Apple's MLX machine learning framework. It links to introductory and deep-dive articles on using MLX for on-device machine learning and LLM deployment on Apple Silicon, models hosted on Hugging Face that are built or fine-tuned for MLX, and a collection of libraries and tools built on top of MLX.

The listed tools cover quantization (a measured-cost quantizer that solves bit-width and group-size allocation for a given RAM budget), running and serving local LLMs and vision-language models, fast Whisper transcription, benchmarking MLX operations across Apple Silicon chips and other GPUs, a HuggingFace Transformers-style model interface, terminal-based model runners, image and video generation, CLIP support, guided generation, and text embeddings, as well as a reinforcement learning library and a ResNet implementation.

Website: https://github.com/antranapp/awesome-mlx