iclementine / optimize_softmaxLinks
Optimize softmax in triton in many cases
☆21Updated 11 months ago
Alternatives and similar repositories for optimize_softmax
Users that are interested in optimize_softmax are comparing it to the libraries listed below
Sorting:
- Examples of CUDA implementations by Cutlass CuTe☆214Updated last month
- ☆89Updated 2 months ago
- 使用 cutlass 仓库在 ada 架构上实现 fp8 的 flash attention☆74Updated 11 months ago
- ☆128Updated 8 months ago
- ☆149Updated 6 months ago
- ☆37Updated last year
- Implement Flash Attention using Cute.☆92Updated 7 months ago
- ☆137Updated last year
- ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.☆93Updated 2 months ago
- This project is about convolution operator optimization on GPU, include GEMM based (Implicit GEMM) convolution.☆37Updated 7 months ago
- 使用 cutlass 实现 flash-attention 精简版,具有教学意义☆45Updated 11 months ago
- ☆51Updated 2 weeks ago
- ☆102Updated 4 months ago
- ☆149Updated 7 months ago
- llm theoretical performance analysis tools and support params, flops, memory and latency analysis.☆99Updated 3 weeks ago
- Optimize GEMM with tensorcore step by step☆31Updated last year
- ☆96Updated 10 months ago
- A Easy-to-understand TensorOp Matmul Tutorial☆369Updated 10 months ago
- ⚡️FFPA: Extend FlashAttention-2 with Split-D, achieve ~O(1) SRAM complexity for large headdim, 1.8x~3x↑ vs SDPA.🎉☆194Updated 2 months ago
- A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer☆94Updated 3 weeks ago
- flash attention tutorial written in python, triton, cuda, cutlass☆398Updated 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
- ☆139Updated last year
- Several optimization methods of half-precision general matrix vector multiplication (HGEMV) using CUDA core.☆63Updated 10 months ago
- A tutorial for CUDA&PyTorch☆150Updated 6 months ago
- ☆67Updated 6 months ago
- A lightweight design for computation-communication overlap.☆155Updated last month
- CUDA 8-bit Tensor Core Matrix Multiplication based on m16n16k16 WMMA API☆31Updated last year
- A GPU-optimized system for efficient long-context LLMs decoding with low-bit KV cache.☆56Updated last week
- 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