NVIDIA / NVIDIA/TransformerEngine
[PyTorch] Reduce CPU overhead in attention
Open
@vcherepanov-nv is already working on this.
Since Dec 9, 2025.
attention
- Dominant language
- Python
- Stars
- 3.5k
- Forks
- 831
- Avg merge
- 3d 11h
- Merged PRs (30d)
- 65
Description
GPU kernels are getting faster and faster on newer architectures, leaving CPU overhead the bottleneck for future workflows. It's time to review TE-PyTorch attention code and see if there are optimization opportunities to reduce the CPU overhead.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.