Decoding Attention is specially optimized for MHA, MQA, GQA and MLA using CUDA core for the decoding stage of LLM inference.
☆46Jun 11, 2025Updated 8 months ago
Alternatives and similar repositories for decoding_attention
Users that are interested in decoding_attention are comparing it to the libraries listed below
Sorting:
- Multiple GEMM operators are constructed with cutlass to support LLM inference.☆20Aug 3, 2025Updated 7 months ago
- Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.☆44Feb 27, 2025Updated last year
- A practical way of learning Swizzle☆37Feb 3, 2025Updated last year
- Several optimization methods of half-precision general matrix vector multiplication (HGEMV) using CUDA core.☆73Sep 8, 2024Updated last year
- ☆262Jul 11, 2024Updated last year
- Quantized Attention on GPU☆44Nov 22, 2024Updated last year
- ☆20Dec 24, 2024Updated last year
- Kernel Library Wheel for SGLang☆16Updated this week
- Awesome code, projects, books, etc. related to CUDA☆31Feb 3, 2026Updated last month
- SGEMM optimization with cuda step by step☆21Mar 23, 2024Updated last year
- ☆116May 16, 2025Updated 9 months ago
- Several optimization methods of half-precision general matrix multiplication (HGEMM) using tensor core with WMMA API and MMA PTX instruct…☆526Sep 8, 2024Updated last year
- ☆14Nov 3, 2025Updated 4 months ago
- Examples of CUDA implementations by Cutlass CuTe☆269Jul 1, 2025Updated 8 months ago
- Benchmark tests supporting the TiledCUDA library.☆18Nov 19, 2024Updated last year
- DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling☆21Feb 9, 2026Updated 3 weeks ago
- High performance RMSNorm Implement by using SM Core Storage(Registers and Shared Memory)☆30Jan 22, 2026Updated last month
- Standalone Flash Attention v2 kernel without libtorch dependency☆114Sep 10, 2024Updated last year
- Implement Flash Attention using Cute.☆101Dec 17, 2024Updated last year
- We invite you to visit and follow our new repository at https://github.com/microsoft/TileFusion. TiledCUDA is a highly efficient kernel …☆194Jan 28, 2025Updated last year
- A standalone GEMM kernel for fp16 activation and quantized weight, extracted from FasterTransformer☆96Feb 20, 2026Updated last week
- llama INT4 cuda inference with AWQ☆54Jan 20, 2025Updated last year
- ☆168Feb 5, 2026Updated last month
- ⚡️Write HGEMM from scratch using Tensor Cores with WMMA, MMA and CuTe API, Achieve Peak⚡️ Performance.☆147May 10, 2025Updated 9 months ago
- A collection of memory efficient attention operators implemented in the Triton language.☆288Jun 5, 2024Updated last year
- PyTorch bindings for CUTLASS grouped GEMM.☆144May 29, 2025Updated 9 months ago
- A Easy-to-understand TensorOp Matmul Tutorial☆410Feb 11, 2026Updated 3 weeks ago
- ☆49Apr 15, 2024Updated last year
- 使用 CUDA C++ 实现的 llama 模型推理框架☆64Nov 8, 2024Updated last year
- CUTLASS and CuTe Examples☆132Nov 30, 2025Updated 3 months ago
- ☆105Nov 7, 2024Updated last year
- Yinghan's Code Sample☆365Jul 25, 2022Updated 3 years ago
- DeeperGEMM: crazy optimized version☆74May 5, 2025Updated 9 months ago
- 使用 cutlass 实现 flash-attention 精简版，具有教学意义☆58Aug 12, 2024Updated last year
- Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity☆234Sep 24, 2023Updated 2 years ago
- Persistent dense gemm for Hopper in `CuTeDSL`☆15Aug 9, 2025Updated 6 months ago
- ☆13Jun 18, 2024Updated last year
- A unified programming framework for high and portable performance across FPGAs and GPUs☆11Mar 23, 2025Updated 11 months ago
- Created a simple neural network using C++17 standard and the Eigen library that supports both forward and backward propagation.☆10Jul 27, 2024Updated last year