metal-performance-shaders

Code examples for new APIs of iOS 10.

A JAX backend for Apple Metal Performance Shaders (MPS), enabling GPU-accelerated JAX computations on Apple Silicon.

A few quick scripts focused on testing TensorFlow/PyTorch/Llama 2 on macOS.

VGGT 3D Vision Agent optimized for Apple Silicon with Metal Performance Shaders

Fast image interpolation using Metal

Bringing StableHLO to Apple Silicon

Run compiled CoreML(v1) model using MPSNN

Defensive layer for mlx / mlx-lm / mlx-vlm on Apple Silicon. Prevents IOGPUFamily kernel panics, SIGABRT, and Mac reboots from MLX inference. Includes a community-curated registry of known-panic models (KNOWN_PANIC_MODELS) with hardware contexts, root-cause hypotheses, and verified workarounds.

👨‍💻🔢🖼️Quick example of how Metal Performance Shaders can be used for Image Processing

Base on "tacotron2-jpanese" builded & change

Train neural networks with hardware acceleration in Swift

This repository provides a guide for installing TensorFlow and PyTorch on Mac computers with Apple Silicon.

Metal

🍎 Accelerate 3D reconstructions from images using VGGT, optimized for Apple Silicon with Metal Performance Shaders for fast and efficient processing.

UltraShape 1.0 ported to Apple Silicon (MPS) -- high-fidelity 3D shape generation from images on Mac

Run CUDA-hardcoded PyTorch repos on Apple Silicon (MPS) with zero source edits

Educational neural-network implementations in Rust and Swift for MNIST, CIFAR-10, XOR, manual backprop, CNN/MLP/self-attention paths, plus Python visualization and companion MLX-based Swift workflows.

⚡ The fastest micromagnetic simulator on a Mac — 3.3×–30× faster and 5.9×–26× more energy-efficient than every other simulator that runs on Apple Silicon. Native Metal port of mumax³: unmodified .mx3, validated physics, 84M cells on a fanless MacBook Air.

Part 2 of Undergraduate Thesis Project @ UniBo: An iOS application using Metal Performance Shaders to classify pill images through a VGG16-like ConvNet