FlashInfer: Kernel Library for LLM Serving
-
Updated
Aug 29, 2025 - Cuda
#
attention
Here are 10 public repositories matching this topic...
Quantized Attention achieves speedup of 2-5x and 3-11x compared to FlashAttention and xformers, without lossing end-to-end metrics across language, image, and video models.
cuda triton attention vit quantization video-generation mlsys inference-acceleration efficient-attention llm llm-infra video-generate
-
Updated
Aug 5, 2025 - Cuda
SpargeAttention: A training-free sparse attention that can accelerate any model inference.
attention vit quantization video-generation mlsys inference-acceleration ai-infra vision-transformer sparse-attention llm sageattention
-
Updated
Aug 13, 2025 - Cuda
🤖FFPA: Extend FlashAttention-2 with Split-D, ~O(1) SRAM complexity for large headdim, 1.8x~3x↑🎉 vs SDPA EA.
cuda attention sdpa mla mlsys tensor-cores flash-attention deepseek deepseek-v3 deepseek-r1 fused-mla flash-mla
-
Updated
Aug 8, 2025 - Cuda
Code for the paper "Cottention: Linear Transformers With Cosine Attention"
-
Updated
Oct 19, 2024 - Cuda
Patch-Based Stochastic Attention (efficient attention mecanism)
-
Updated
Jan 16, 2023 - Cuda
easy naive flash attention without optimization base on origin paper
-
Updated
Jun 25, 2025 - Cuda
A simple implementation of PagedAttention purely written in CUDA and C++.
-
Updated
Aug 24, 2025 - Cuda
Improve this page
Add a description, image, and links to the attention topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the attention topic, visit your repo's landing page and select "manage topics."