thu-ml / SpargeAttn
[ICML2025] SpargeAttention: A training-free sparse attention that accelerates any model inference.
Why this job: topics: video-generation
What it does
We highly recommend using the spas_sage2_attn_meansim_topk_cuda and block_sparse_sage2_attn_cuda APIs. They are plug-and-play and customizable:
Get started
pip install ninja
# for parallel compilation
python setup.py install
# or pip install -e .- Maintenance
- Inactive
- Last commit 8 months ago
- Stars
- 1,267
- +1 this week
- Licence
- Apache-2.0
- Allows business use
- Downloads for
- None listed
- No release files
- Latest version
- None yet
- No releases
- Built with
- Cuda
- 59% of the code
Is it looked after?
- Contributors
- Not stated
- Last 90 days
- Issues
- 39% closed
- 63 still open
- Pull requests
- 10 merged
- 4 waiting
- Releases
- None
- Never published
- No contributing guide
- No security policy
- No code of conduct
More: languages, topics and project record
Languages
- Cuda 59%
- Python 39%
- C 1%
- C++ 1%
Topics
- ai-infra
- attention
- inference-acceleration
- llm
- mlsys
- quantization
- sparse-attention
- vision-transformer
- sageattention
- video-generation
- vit
Project record
- JobVideo
- StatusListed
- Commits110
- Forks108
- Watching10
- Stars in 28 days+8
- CreatedFeb 25, 2025
- First seen by usNov 7, 2026
- CheckedNov 8, 2026