spb/tools-spboucher-ai

1{"slug":"apple-silicon-ai","title":"Apple Silicon AI","published":true,2 "description":"Running and serving language, speech and vision models on Macs with M-series chips, from the array framework up to the applications that wrap it. Useful for anyone building a local or cluster-based inference setup on macOS rather than renting GPUs.",3 "items":[4  {"tool":"mlx","blurb":"Apple's array framework with unified-memory semantics; the foundation for most native Apple Silicon ML work."},5  {"tool":"mlx-lm","blurb":"Generation, quantization and fine-tuning of language models on top of MLX, with Hugging Face Hub integration."},6  {"tool":"mlx-whisper","blurb":"Speech-to-text with Whisper models ported to MLX; practical for transcribing lectures and meetings locally."},7  {"tool":"llama-cpp","blurb":"Portable C/C++ inference with Metal acceleration and the GGUF format; the other mature path for local LLMs on a Mac."},8  {"tool":"ollama","blurb":"The simplest way to pull and serve open models locally behind an HTTP API."},9  {"tool":"lm-studio","blurb":"Desktop application for discovering, downloading and chatting with local models."},10  {"tool":"core-ml","blurb":"Apple's on-device inference framework for shipping models inside macOS and iOS apps."},11  {"tool":"apple-foundation-models","blurb":"The system language model exposed to Swift apps on Apple platforms."},12  {"tool":"hugging-face-hub","blurb":"Where the model weights come from; MLX and GGUF conversions are published there."},13  {"tool":"exo","blurb":"Experimental distributed inference across several Apple devices."}14 ]}15