flashmla

Here are 4 public repositories matching this topic...

Bruce-Lee-LY / decoding_attention

Decoding Attention is specially optimized for MHA, MQA, GQA and MLA using CUDA core for the decoding stage of LLM inference.

gpu cuda inference nvidia mha mla multi-head-attention gqa mqa llm large-language-model flash-attention cuda-core decoding-attention flashinfer flashmla

Updated Jun 11, 2025
C++

kamalrss88 / FlashMLA

Star

🚀 Accelerate attention mechanisms with FlashMLA, featuring optimized kernels for DeepSeek models, enhancing performance through sparse and dense attention.

windows gpu cuda inference nvidia nvidia-cuda mla multi-head-attention mqa llm flash-attention cuda-core decoding-attention deepseek flashinfer flashmla

Updated Aug 13, 2026
C++

Cat-Gawr / DeepSeek-FlashMLA

Star

DeepSeek Flash MLA - DeepSeek - copy manual

windows nvidia-cuda deepseek flashmla

Updated Apr 22, 2025
C++

aymanelrody / FlashMLA

Star

⚡ Optimize attention mechanisms with FlashMLA, a library of advanced sparse and dense kernels for DeepSeek models, improving performance and efficiency.

windows gpu cuda inference nvidia mha mla multi-head-attention gqa mqa llm flash-attention cuda-core decoding-attention deepseek flashinfer flashmla

Updated Aug 13, 2026
C++

Improve this page

Add a description, image, and links to the flashmla topic page so that developers can more easily learn about it.

Curate this topic

Add this topic to your repo

To associate your repository with the flashmla topic, visit your repo's landing page and select "manage topics."

Learn more