microsoft / TileFusionLinks
TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.
☆100Updated 3 months ago
Alternatives and similar repositories for TileFusion
Users that are interested in TileFusion 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 …☆186Updated 8 months ago
- Framework to reduce autotune overhead to zero for well known deployments.☆84Updated last month
- ☆65Updated 6 months ago
- ☆46Updated 5 months ago
- ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.☆121Updated 5 months ago
- DeeperGEMM: crazy optimized version☆72Updated 5 months ago
- ☆92Updated 11 months ago
- High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.☆117Updated last year
- Standalone Flash Attention v2 kernel without libtorch dependency☆112Updated last year
- Github mirror of trition-lang/triton repo.☆86Updated last week
- Multi-Level Triton Runner supporting Python, IR, PTX, and cubin.☆73Updated last week
- Implement Flash Attention using Cute.☆96Updated 10 months ago
- FractalTensor is a programming framework that introduces a novel approach to organizing data in deep neural networks (DNNs) as a list of …☆29Updated 10 months ago
- extensible collectives library in triton☆90Updated 6 months ago
- Tile-based language built for AI computation across all scales☆67Updated 3 weeks ago
- NVSHMEM‑Tutorial: Build a DeepEP‑like GPU Buffer☆139Updated last month
- A Triton JIT runtime and ffi provider in C++☆27Updated this week
- GPTQ inference TVM kernel☆39Updated last year
- A GPU-optimized system for efficient long-context LLMs decoding with low-bit KV cache.☆60Updated this week
- A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer