pawel-mazurkiewicz/ComfyUI-AppleSilicon-FP8

Apple Silicon acceleration for ComfyUI, featuring custom Metal kernels, faster image and video generation, and FP8/INT8 model support on Mac.

Real-time multimodal AI pipelines on Apple Silicon

works well with rocks; native Apple Silicon inference engine for capable and tiny models (ternary, Qwen) via MLX Swift. Hadamard-rotated KV cache, vision, tool calling, OpenAI + Anthropic APIs

JANG — GGUF for MLX. YOU MUST USE JANG_Q RUNTIME. Adaptive Mixed-Precision Quantization + Runtime for Apple Silicon

Fork of antirez/ds4 for Apple Silicon, tuned mostly for M1/M2: Qwen3.8-Flash-Next with 262K context on 64 GB and up, 35 tok/s on an M1 Max, MTP and vision, one-line install.

W8A8/W4A8 inference + optimized SDPA on Apple Silicon — unlocking unused INT8 TensorOps in M5 for 1.2–1.9× faster LLM prefill, plus FlashInfer-inspired GQA decode attention for up to 1.6× SDPA speedup, built as MLX custom primitives.

Stable Diffusion with Core ML on Apple Silicon

fastest runtime for apple silicon.

⚡ Native MLX Swift LLM inference server for Apple Silicon. OpenAI-compatible API, SSD streaming for 100B+ MoE models, TurboQuant KV cache compression, MACOS + iOS iPhone app.