ARM-software / kleidiaiLinks
This repository is a read-only mirror of https://gitlab.arm.com/kleidi/kleidiai
☆91Updated this week
Alternatives and similar repositories for kleidiai
Users that are interested in kleidiai are comparing it to the libraries listed below
Sorting:
- FlagTree is a unified compiler for multiple AI chips, which is forked from triton-lang/triton.☆93Updated this week
- ☆167Updated last week
- Benchmark code for the "Online normalizer calculation for softmax" paper☆102Updated 7 years ago
- OpenAI Triton backend for Intel® GPUs☆212Updated last week
- BitBLAS is a library to support mixed-precision matrix multiplications, especially for quantized LLM deployment.☆703Updated 2 months ago
- 🤖FFPA: Extend FlashAttention-2 with Split-D, ~O(1) SRAM complexity for large headdim, 1.8x~3x↑🎉 vs SDPA EA.☆226Updated 2 months ago
- ☆163Updated 2 years ago
- triton for dsa☆43Updated last week
- High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.☆117Updated last year
- ☆176Updated 2 years ago
- An experimental CPU backend for Triton☆154Updated last week
- ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.☆124Updated 5 months ago
- PyTorch emulation library for Microscaling (MX)-compatible data formats☆307Updated 4 months ago
- Several optimization methods of half-precision general matrix vector multiplication (HGEMV) using CUDA core.☆67Updated last year
- ☆82Updated 9 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 9 months ago
- ☆74Updated 11 months ago
- Composable Kernel: Performance Portable Programming Model for Machine Learning Tensor Operators☆478Updated this week
- PyTorch extension for emulating FP8 data formats on standard FP32 Xeon/GPU hardware.☆111Updated 10 months ago
- This is a demo how to write a high performance convolution run on apple silicon☆56Updated 3 years ago
- An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).☆266Updated 3 months ago
- llama INT4 cuda inference with AWQ☆55Updated 9 months ago
- Standalone Flash Attention v2 kernel without libtorch dependency☆112Updated last year
- A quantization algorithm for LLM☆143Updated last year
- ☆91Updated this week
- QQQ is an innovative and hardware-optimized W4A8 quantization solution for LLMs.☆144Updated 2 months ago
- ☆241Updated last year
- ☆109Updated 6 months ago
- CUDA Matrix Multiplication Optimization☆234Updated last year
- ☆46Updated last year