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 supported18 Subscribers
Add a CodeTriage badge to deepspeed
Help out
- Issues
- Stop obsolete CI jobs on workflow cancellation
- Enable bf16 check_grad_overflow by default (matching fp16)
- [Draft] Add ZeRO-3 elastic checkpoint save/load support
- [Draft] Add On-Policy Distillation (OPSD) Trainer in DeepSpeed
- EvoformerAttention should auto-detect CUTLASS instead of requiring CUTLASS_PATH
- [REQUEST]Will zero 3 support diffrent module usage?
- [BUG]Loss connection to ranks
- Incorrect variable name: `self.ev_values`
- Add Qwen 3.5 preset to AutoTP
- [REQUEST] Add Trackio as a New Backend for Experiment Monitoring
- Docs
- Python not yet supported