Repeerc / flash-attention-v2-RDNA3-minimalLinks
a simple Flash Attention v2 implementation with ROCM (RDNA3 GPU, roc wmma), mainly used for stable diffusion(ComfyUI) in Windows ZLUDA environments.
☆48Updated last year
Alternatives and similar repositories for flash-attention-v2-RDNA3-minimal
Users that are interested in flash-attention-v2-RDNA3-minimal are comparing it to the libraries listed below
Sorting:
- Fast and memory-efficient exact attention☆191Updated this week
- Model Compression Toolbox for Large Language Models and Diffusion Models☆655Updated last month
- ☆76Updated 9 months ago
- QQQ is an innovative and hardware-optimized W4A8 quantization solution for LLMs.☆141Updated last month
- QuTLASS: CUTLASS-Powered Quantized BLAS for Deep Learning☆111Updated last week
- A CUDA kernel for NHWC GroupNorm for PyTorch☆21Updated 10 months ago
- ☆56Updated 2 months ago
- ☆152Updated 3 months ago
- Development repository for the Triton language and compiler☆133Updated this week
- An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).☆265Updated 2 months ago
- ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.☆119Updated 5 months ago
- High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.☆116Updated last year
- A safetensors extension to efficiently store sparse quantized tensors on disk☆167Updated this week
- ☆82Updated 8 months ago
- 使用 cutlass 仓库在 ada 架构上实现 fp8 的 flash attention☆75Updated last year
- AI Tensor Engine for ROCm☆284Updated this week
- High performance inference engine for diffusion models☆93Updated last month
- ☆64Updated 5 months ago
- 🎬 3.7× faster video generation E2E 🖼️ 1.6× faster image generation E2E ⚡ ColumnSparseAttn 9.3× vs FlashAttn‑3 💨 ColumnSparseGEMM 2.5× …☆86Updated last month
- Standalone Flash Attention v2 kernel without libtorch dependency☆111Updated last year
- Several optimization methods of half-precision general matrix vector multiplication (HGEMV) using CUDA core.☆67Updated last year
- Ahead of Time (AOT) Triton Math Library☆77Updated this week
- We invite you to visit and follow our new repository at https://github.com/microsoft/TileFusion. TiledCUDA is a highly efficient kernel …☆186Updated 8 months ago
- 🤖FFPA: Extend FlashAttention-2 with Split-D, ~O(1) SRAM complexity for large headdim, 1.8x~3x↑🎉 vs SDPA EA.☆220Updated 2 months ago
- A GPU-optimized system for efficient long-context LLMs decoding with low-bit KV cache.☆60Updated last month
- ☆115Updated last month
- Implement Flash Attention using Cute.☆96Updated 9 months ago
- ☆99Updated 4 months ago
- ☆33Updated 8 months ago
- Fast Hadamard transform in CUDA, with a PyTorch interface☆245Updated last week