A fast communication-overlapping library for tensor parallelism on GPUs.

🔥🔥🔥 A collection of some awesome public CUDA, cuBLAS, cuDNN, CUTLASS, TensorRT, TensorRT-LLM, Triton, MLIR and High Performance Computing (HPC) projects.

Examples of CUDA implementations by Cutlass CuTe

Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.

GEMM and Winograd based convolutions using CUTLASS

CUTLASS and CuTe Examples

study of cutlass

Multiple GEMM operators are constructed with cutlass to support LLM inference.

pytorch implements block sparse