Gemm Python Packages

ctranslate2

Fast inference engine for Transformer models

nvidia-cudnn-frontend

cuDNN Frontend is NVIDIA's modern, open-source entry point to the cuDNN library and a growing collection of high-performance open-source kernels.

nanogemm

Minimalist, bare-metal SIMD & Assembly GEMM engine for Python. Sub-microsecond CPU matrix multiplication & Native MCP Server (nanogemm-mcp).

tokenspeed-iris

AMD RAD's multi-GPU Triton-based framework for seamless multi-GPU programming

voxrun-ctranslate2

Fast inference engine for Transformer models

ctranslate2-arty

Fast inference engine for Transformer models

trnblas

BLAS Levels 1–3 for AWS Trainium via NKI (cuBLAS-equivalent) — GEMM with stationary-tile reuse, batched GEMM, TRSM, validated DF-MP2 for quantum chemistry.

pyclblast

Tuned OpenCL BLAS

sparsemma

INT8 Sparse Tensor Core GEMM for PyTorch — built for Windows