seb-v / fp32_sgemm_amdLinks
Super fast FP32 matrix multiplication on RDNA3
☆71Updated 4 months ago
Alternatives and similar repositories for fp32_sgemm_amd
Users that are interested in fp32_sgemm_amd are comparing it to the libraries listed below
Sorting:
- High-Performance SGEMM on CUDA devices☆97Updated 7 months ago
- A tool for generating information about the matrix multiplication instructions in AMD Radeon™ and AMD Instinct™ accelerators☆112Updated 3 months ago
- Unofficial description of the CUDA assembly (SASS) instruction sets.☆136Updated last month
- An experimental CPU backend for Triton (https//github.com/openai/triton)☆44Updated last week
- Unified compiler/runtime for interfacing with PyTorch Dynamo.☆101Updated last week
- GPUOcelot: A dynamic compilation framework for PTX☆207Updated 6 months ago
- rocWMMA☆125Updated this week
- ☆147Updated this week
- Efficient implementation of DeepSeek Ops (Blockwise FP8 GEMM, MoE, and MLA) for AMD Instinct MI300X☆67Updated 3 weeks ago
- Nvidia Instruction Set Specification Generator☆290Updated last year
- Tenstorrent MLIR compiler☆174Updated this week
- AI Tensor Engine for ROCm☆260Updated this week
- Ahead of Time (AOT) Triton Math Library☆75Updated this week
- Official Problem Sets / Reference Kernels for the GPU MODE Leaderboard!☆74Updated this week
- An interactive web-based tool for exploring intermediate representations of PyTorch and Triton models☆48Updated last week
- IREE's PyTorch Frontend, based on Torch Dynamo.☆95Updated this week
- ☆58Updated this week
- Development repository for the Triton language and compiler☆127Updated this week
- MLIR-based partitioning system☆123Updated last week
- amdgpu example code in hip/asm☆39Updated last week
- Attention in SRAM on Tenstorrent Grayskull☆38Updated last year
- CUDA Matrix Multiplication Optimization☆218Updated last year
- A simple profiler to count Nvidia PTX assembly instructions of OpenCL/SYCL/CUDA kernels for roofline model analysis.☆55Updated 5 months ago
- TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.☆95Updated 2 months ago
- Tilus is a tile-level kernel programming language with explicit control over shared memory and registers.☆317Updated this week
- ☆42Updated 3 months ago
- TritonParse: A Compiler Tracer, Visualizer, and mini-Reproducer(WIP) for Triton Kernels☆144Updated this week
- An extension library of WMMA API (Tensor Core API)☆103Updated last year
- High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.☆113Updated last year
- ☆49Updated 7 months ago