xlite-dev / HGEMMLinks
⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.
☆91Updated 2 months ago
Alternatives and similar repositories for HGEMM
Users that are interested in HGEMM are comparing it to the libraries listed below
Sorting:
- We invite you to visit and follow our new repository at https://github.com/microsoft/TileFusion. TiledCUDA is a highly efficient kernel …☆183Updated 6 months ago
- Implement Flash Attention using Cute.☆89Updated 7 months ago
- ☆89Updated 2 months ago
- 使用 cutlass 仓库在 ada 架构上实现 fp8 的 flash attention☆74Updated 11 months ago
- A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer☆93Updated 2 weeks ago
- ☆96Updated 10 months ago
- ☆60Updated 3 months ago
- Standalone Flash Attention v2 kernel without libtorch dependency☆111Updated 10 months ago
- ⚡️FFPA: Extend FlashAttention-2 with Split-D, achieve ~O(1) SRAM complexity for large headdim, 1.8x~3x↑ vs SDPA.🎉☆192Updated 2 months ago
- Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.☆39Updated 5 months ago
- ☆37Updated last year
- Several optimization methods of half-precision general matrix vector multiplication (HGEMV) using CUDA core.☆63Updated 10 months ago
- A lightweight design for computation-communication overlap.☆154Updated last month
- Examples of CUDA implementations by Cutlass CuTe☆211Updated last month
- ☆31Updated 5 months ago
- A GPU-optimized system for efficient long-context LLMs decoding with low-bit KV cache.☆56Updated this week
- ☆51Updated 2 weeks ago
- DeeperGEMM: crazy optimized version☆70Updated 2 months ago
- Decoding Attention is specially optimized for MHA, MQA, GQA and MLA using CUDA core for the decoding stage of LLM inference.☆40Updated last month
- Triton adapter for Ascend. Mirror of https://gitee.com/ascend/triton-ascend☆60Updated last week
- llama INT4 cuda inference with AWQ☆54Updated 6 months ago
- ☆102Updated 4 months ago
- TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.☆91Updated last month
- ☆128Updated 7 months ago
- ☆227Updated last year
- play gemm with tvm☆91Updated 2 years ago
- Benchmark code for the "Online normalizer calculation for softmax" paper☆95Updated 7 years ago
- SpInfer: Leveraging Low-Level Sparsity for Efficient Large Language Model Inference on GPUs☆50Updated 4 months ago
- Optimize GEMM with tensorcore step by step☆31Updated last year
- High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.☆113Updated last year