Bruce-Lee-LY / cutlass_gemmLinks
Multiple GEMM operators are constructed with cutlass to support LLM inference.
☆19Updated 2 months ago
Alternatives and similar repositories for cutlass_gemm
Users that are interested in cutlass_gemm are comparing it to the libraries listed below
Sorting:
- Decoding Attention is specially optimized for MHA, MQA, GQA and MLA using CUDA core for the decoding stage of LLM inference.☆44Updated 3 months ago
- 使用 cutlass 仓库在 ada 架构上实现 fp8 的 flash attention☆75Updated last year
- ☆64Updated 5 months ago
- ☆13Updated 7 months ago
- ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.☆118Updated 4 months ago
- Standalone Flash Attention v2 kernel without libtorch dependency☆110Updated last year
- A practical way of learning Swizzle☆28Updated 8 months ago
- ☆32Updated 8 months ago
- ☆17Updated last year
- ☆56Updated 2 months ago
- Persistent dense gemm for Hopper in `CuTeDSL`☆15Updated last month
- Tutorials of Extending and importing TVM with CMAKE Include dependency.☆14Updated 11 months ago
- High Performance FP8 GEMM Kernels for SM89 and later GPUs.☆20Updated 8 months ago
- 使用 cutlass 实现 flash-attention 精简版,具有教学意义☆49Updated last year
- Implement Flash Attention using Cute.☆96Updated 9 months ago
- ☆98Updated last year
- 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
- A Triton JIT runtime and ffi provider in C++☆25Updated 2 weeks ago
- 使用 CUDA C++ 实现的 llama 模型推理框架☆62Updated 10 months ago
- ☆43Updated last year
- GPTQ inference TVM kernel☆40Updated last year
- study of cutlass☆22Updated 10 months ago
- ☆106Updated 4 months ago
- Quantized Attention on GPU☆44Updated 10 months ago
- Awesome code, projects, books, etc. related to CUDA☆24Updated last month
- DeeperGEMM: crazy optimized version☆71Updated 5 months ago
- Several optimization methods of half-precision general matrix vector multiplication (HGEMV) using CUDA core.☆67Updated 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
- An experimental communicating attention kernel based on DeepEP.☆34Updated 2 months ago