simveit / persistent_dense_gemmLinks
Persistent dense gemm for Hopper in `CuTeDSL`
☆15Updated 2 weeks ago
Alternatives and similar repositories for persistent_dense_gemm
Users that are interested in persistent_dense_gemm are comparing it to the libraries listed below
Sorting:
- A practical way of learning Swizzle☆25Updated 6 months ago
- High Performance FP8 GEMM Kernels for SM89 and later GPUs.☆17Updated 6 months ago
- Multiple GEMM operators are constructed with cutlass to support LLM inference.☆19Updated 2 weeks ago
- Triton adapter for Ascend. Mirror of https://gitee.com/ascend/triton-ascend☆64Updated this week
- ☆61Updated 3 months ago
- Decoding Attention is specially optimized for MHA, MQA, GQA and MLA using CUDA core for the decoding stage of LLM inference.☆40Updated 2 months ago
- Implement Flash Attention using Cute.☆94Updated 8 months ago
- ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.☆100Updated 3 months ago
- SpInfer: Leveraging Low-Level Sparsity for Efficient Large Language Model Inference on GPUs☆52Updated 4 months ago
- Quantized Attention on GPU☆44Updated 9 months ago
- 使用 cutlass 仓库在 ada 架构上实现 fp8 的 flash attention☆74Updated last year
- ☆12Updated 5 months ago
- ☆50Updated 3 months ago
- ☆39Updated last year
- TileFusion is an experimental C++ macro kernel template library that elevates the abstraction level in CUDA C for tile processing.☆94Updated last month
- Standalone Flash Attention v2 kernel without libtorch dependency☆111Updated 11 months ago
- GPTQ inference TVM kernel☆40Updated last year
- ☆77Updated 3 months ago
- ☆91Updated this week
- ☆19Updated 10 months ago
- study of cutlass☆22Updated 9 months ago
- A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer☆94Updated 2 weeks ago
- Tutorials of Extending and importing TVM with CMAKE Include dependency.☆14Updated 10 months ago
- Benchmark tests supporting the TiledCUDA library.☆17Updated 9 months ago
- ☆81Updated 7 months ago
- Several optimization methods of half-precision general matrix vector multiplication (HGEMV) using CUDA core.☆63Updated 11 months ago
- A GPU-optimized system for efficient long-context LLMs decoding with low-bit KV cache.☆56Updated 2 weeks ago
- DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling☆18Updated this week
- Awesome code, projects, books, etc. related to CUDA☆23Updated last week
- Framework to reduce autotune overhead to zero for well known deployments.☆80Updated last week