⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.
☆149May 10, 2025Updated 10 months ago
Alternatives and similar repositories for HGEMM
Users that are interested in HGEMM are comparing it to the libraries listed below
Sorting:
- 🤖FFPA: Extend FlashAttention-2 with Split-D, ~O(1) SRAM complexity for large headdim, 1.8x~3x↑🎉 vs SDPA EA.☆253Feb 13, 2026Updated last month
- ☆65Apr 26, 2025Updated 10 months ago
- 使用 cutlass 仓库在 ada 架构上实现 fp8 的 flash attention☆80Aug 12, 2024Updated last year
- Benchmark tests supporting the TiledCUDA library.☆18Nov 19, 2024Updated last year
- ☆20Sep 28, 2024Updated last year
- We invite you to visit and follow our new repository at https://github.com/microsoft/TileFusion. TiledCUDA is a highly efficient kernel …☆192Jan 28, 2025Updated last year
- Quantized Attention on GPU☆44Nov 22, 2024Updated last year
- ☆52May 19, 2025Updated 10 months ago
- Examples of CUDA implementations by Cutlass CuTe☆270Jul 1, 2025Updated 8 months ago
- Implement Flash Attention using Cute.☆102Dec 17, 2024Updated last year
- flash attention tutorial written in python, triton, cuda, cutlass☆491Jan 20, 2026Updated 2 months ago
- A CUDA kernel for NHWC GroupNorm for PyTorch☆22Nov 15, 2024Updated last year
- A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer☆95Feb 20, 2026Updated last month
- High performance RMSNorm Implement by using SM Core Storage(Registers and Shared Memory)☆29Jan 22, 2026Updated last month
- ☆261Jul 11, 2024Updated last year
- ☆33Feb 3, 2025Updated last year
- Several optimization methods of half-precision general matrix multiplication (HGEMM) using tensor core with WMMA API and MMA PTX instruct…☆530Sep 8, 2024Updated last year
- An experimental communicating attention kernel based on DeepEP.☆35Jul 29, 2025Updated 7 months ago
- TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.☆106Jun 28, 2025Updated 8 months ago
- TiledLower is a Dataflow Analysis and Codegen Framework written in Rust.☆13Nov 23, 2024Updated last year
- ☆182May 7, 2025Updated 10 months ago
- Multiple GEMM operators are constructed with cutlass to support LLM inference.☆19Aug 3, 2025Updated 7 months ago
- DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling☆21Updated this week
- [HPCA 2026] A GPU-optimized system for efficient long-context LLMs decoding with low-bit KV cache.☆81Dec 18, 2025Updated 3 months ago
- ☆39Dec 14, 2025Updated 3 months ago
- ☆116May 16, 2025Updated 10 months ago
- ☆158Dec 26, 2024Updated last year
- A Easy-to-understand TensorOp Matmul Tutorial☆409Mar 5, 2026Updated 2 weeks ago
- BitBLAS is a library to support mixed-precision matrix multiplications, especially for quantized LLM deployment.☆752Aug 6, 2025Updated 7 months ago
- ☆119May 19, 2025Updated 10 months ago
- DeeperGEMM: crazy optimized version☆75May 5, 2025Updated 10 months ago
- A practical way of learning Swizzle☆37Feb 3, 2025Updated last year
- ☆32Jul 2, 2025Updated 8 months ago
- Cute layout visualization☆32Jan 18, 2026Updated 2 months ago
- Decoding Attention is specially optimized for MHA, MQA, GQA and MLA using CUDA core for the decoding stage of LLM inference.☆45Jun 11, 2025Updated 9 months ago
- Distributed Compiler based on Triton for Parallel Systems☆1,386Mar 11, 2026Updated last week
- ☆137Aug 18, 2025Updated 7 months ago
- PyTorch bindings for CUTLASS grouped GEMM.☆146May 29, 2025Updated 9 months ago
- ☆150Jan 9, 2025Updated last year