deepspeed
https://github.com/microsoft/deepspeed
Python
DeepSpeed is a deep learning optimization library that makes distributed training easy, efficient, and effective.
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported21 Subscribers
Add a CodeTriage badge to deepspeed
Help out
- Issues
- Fix: [Bug] AutoEP FP32 gradient clipping averages norms across ranks with d
- Gradient scale under sequence parallelism differs between ZeRO reduction paths
- fix(checkpoint): support bf16 stage 0 conversion
- fix(checkpointing): reject deallocated input reuse
- fix(pipe): preserve tensor gradient metadata
- Apply the ZeRO-3 predivide postscale when the factor equals world size
- Apply weight decay to the CPU Adagrad update
- DeepSpeed - Q4 2026 Roadmap
- Fuse mixed-precision gradients and parameter writeback into FusedAdam
- Reject Adam settings SuperOffload silently replaces with AdamW
- Docs
- Python not yet supported