Metal Python Packages

wgpu

WebGPU for Python

quadrants

High-performance multi-platform compiler for physics simulation

pygfx

Powerful and versatile visualization for Python.

apache-tvm

Open Machine Learning Compiler Framework

gstaichi

High-performance multi-platform compiler for physics simulation

squish-ai

⚡️ The fastest way to run local LLMs on Apple Silicon — sub-second model loads, beats Ollama on throughput, tail latency, and full-response time. OpenAI/Ollama-compatible. No cloud, no API keys.

pyclesperanto

GPU-accelerated Image Processing library for Python

ttnn

:metal: TT-NN operator library, and TT-Metalium low level kernel programming model.

llama-cpp-py-sync

Auto-synced CFFI ABI python bindings for llama.cpp with prebuilt wheels (CPU/CUDA/Vulkan/Metal).

mtplx

2.24x decode TPS increase On Qwen 3.6 27B @ temp 0.6 | Native MTP Speculative Decoding On Apple Silicon With No External Drafter.

mlx-lattice

Sparse convolution library for MLX designed for point cloud operation.

uzu

A high-performance inference engine for AI models

opencode-llmstack

Cursor-Auto / Claude-tier-style serving for local GGUF models on Mac (M4 Max, 64 GB). FastAPI router fronts llama-swap + llama.cpp, classifying each request into a coder, planner, or uncensored-planner tier. OpenAI-compatible API, opencode integration, per-project subshell, one `llmstack` console-script.

ogre-python

high-performance rendering backend (C++, Python, C#, Java)

tridec

Vendor-portable GPU decoders for quantum LDPC codes — Triton min-sum BP & Relay-BP on NVIDIA (CUDA), AMD (ROCm), and Apple-silicon (Metal), consuming any stim DetectorErrorModel.

mlx-quant-fidelity

Measure MLX quantization quality loss — KL divergence, perplexity, top-token agreement for KV cache and weights

demucs-mlx

Split any song into stems — vocals, drums, bass, and more. Fast music source separation for Apple Silicon, powered by MLX.

xtc-tvm-python-bindings

Open Machine Learning Compiler Framework

alloy-kit

Kernel authoring DSL, torch.compile backend and LLM serving for Apple Silicon.

phi-3-vision-mlx

Phi-3.5 for Mac: Locally-run Vision and Language Models for Apple Silicon

mlx-spectro

High-performance STFT/iSTFT for Apple MLX with fused Metal kernels and autograd support

metalgpu

A python interface for Apple GPU's Metal API

lightning-core

Lightning Core: macOS-first CUDA-style runtime with Metal backend

mhc-mlx

MLX + Metal implementation of mHC: Manifold-Constrained Hyper-Connections by DeepSeek-AI.