Add torch grouped gemm bf16 and mxfp8 support w/ cuda graphed + inference_optimized MoEs#3858
Merged
sidsingh-nvidia merged 56 commits intoNVIDIA:mainfrom Mar 17, 2026
Commits
Commits on Mar 13, 2026
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- andauthored
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
- committed
![claude[bot]](https://avatars.githubusercontent.com/in/1236702?v=4&size=32){kind=avatar}
Commits on Mar 16, 2026
- authored
- committed
- committed
- committed
- committed
- committed
- committed
- authored
- committed
- committed
- committed
- committed
- authored
- committed
- committed
- committed
- committed
Commits on Mar 17, 2026
- committed
- authored
- committed
- committed
- committed
- authored
- committed
- committed
- authored
- committed
- committed