10 projects
mps-flash-attn
Flash Attention for PyTorch on Apple Silicon (M1/M2/M3/M4)
metal-profiler
Static analysis profiler for Metal compute shaders on Apple Silicon
metal-debug
Printf-style debugging for Metal compute shaders
mps-linear-attn
Flash Linear Attention (DeltaNet) for PyTorch on Apple Silicon (MPS)
mps-bitsandbytes
NF4/FP4/FP8/INT8 quantization for PyTorch on Apple Silicon with Metal GPU acceleration
mps-deform-conv
Deformable Convolution 2D for PyTorch on Apple Silicon (MPS)
mps-carafe
CARAFE content-aware upsampling for Apple Silicon (MPS)
mps-correlation
Correlation layer for optical flow on Apple Silicon (MPS)
mps-conv3d
3D Convolution for Apple Silicon (MPS)
mpsops
CUDA-only PyTorch operations for Apple Silicon - metapackage
