wangsiping97 / FastGEMVLinks
High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.
☆116Updated last year
Alternatives and similar repositories for FastGEMV
Users that are interested in FastGEMV are comparing it to the libraries listed below
Sorting:
- Standalone Flash Attention v2 kernel without libtorch dependency☆110Updated last year
- Several optimization methods of half-precision general matrix vector multiplication (HGEMV) using CUDA core.☆67Updated last year
- llama INT4 cuda inference with AWQ☆55Updated 8 months ago
- An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).☆265Updated 2 months ago
- Benchmark code for the "Online normalizer calculation for softmax" paper☆101Updated 7 years ago
- ☆238Updated last year
- TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.☆97Updated 3 months ago
- A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer☆94Updated 3 weeks ago
- Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.☆40Updated 7 months ago
- ☆82Updated 8 months ago
- QuTLASS: CUTLASS-Powered Quantized BLAS for Deep Learning☆99Updated 3 weeks ago
- ☆90Updated 10 months ago
- PyTorch bindings for CUTLASS grouped GEMM.☆123Updated 4 months ago
- 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
- ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.☆118Updated 4 months ago
- ☆98Updated last year
- 使用 cutlass 仓库在 ada 架构上实现 fp8 的 flash attention☆75Updated last year
- Implement Flash Attention using Cute.☆95Updated 9 months ago
- ☆64Updated 5 months ago
- extensible collectives library in triton☆88Updated 6 months ago
- DeeperGEMM: crazy optimized version☆71Updated 4 months ago
- Fast Hadamard transform in CUDA, with a PyTorch interface☆239Updated last month
- A GPU-optimized system for efficient long-context LLMs decoding with low-bit KV cache.☆60Updated last month
- ☆121Updated 9 months ago
- ☆159Updated 2 years ago
- ☆151Updated last year
- QQQ is an innovative and hardware-optimized W4A8 quantization solution for LLMs.☆140Updated last month
- ☆106Updated 4 months ago
- ☆43Updated last year
- Ahead of Time (AOT) Triton Math Library☆77Updated this week