#efficient-attention Repositories
Discover 3 active open source GitHub repositories tagged with #efficient-attention, accumulating 5,099 total stars (averaging 1,699 stars per project). Leading this topic is thu-ml/SageAttention with 4,017 stars.
3 repositories total
[ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-to-end metrics across language, image, and video models.
Implementation of 💍 Ring Attention, from Liu et al. at Berkeley AI, in Pytorch
[IEEE TMI-2024] UNETR++: Delving into Efficient and Accurate 3D Medical Image Segmentation
Frequently Asked Questions
What is the most popular #efficient-attention repository on GitHub?
The top ranked open source #efficient-attention project on GitHub is thu-ml/SageAttention with 4,017 stars. [ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-to-end metrics across language, image, and video models.
Which programming languages are most used in #efficient-attention repositories?
Among the top ranked #efficient-attention projects on GitHub, the most common programming languages are Python, Cuda.
How is the #efficient-attention repository ranking calculated?
We track 3 active open source repositories tagged with #efficient-attention (500+ stars), totaling 5,099 stars with an average of 1,699 stars per repository. Rankings update daily from the GitHub API.