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
- docs(inference): align fp_quantize's docstring with its signature
- docs: remove duplicated words and garbled fragments in comments and docs
- Weight the TiledLoss backward pass the way the forward does
- init_inference rejects dtype=torch.int8 on every accelerator
- [muon] Per-head Newton-Schulz for fused QKV projections
- Release AutoEP router-logit caches on checkpoint replay exits
- [AutoTP] Implement VocabParallelEmbedding and enhance tied model support
- fix(rollout): reject shared prefill with continuous batching
- Add `DualPipeV`
- Accept torch.int8 in init_inference's dtype validation gate
- Docs
- Python not yet supported