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 supported15 Subscribers
Add a CodeTriage badge to deepspeed
Help out
- Issues
- [BUG] z3+compile+gradient checkpoint uses more memory
- Enabled high-performance Automatic Tensor Parallelism (auto TP) for the MoE models on multiple GPUs/HPUs
- [REQUEST] FPDT backward test
- [FPDT] Support FPDT Based on Intel Backend
- Is "Hierarchical All-to-all" feat available in current version?
- [REQUEST] Pipeline Parallelism support multi optimizer to train
- Update sharded_moe.py to support top2 gate with Tutel
- [BUG] `deepspeed.initialize` changes the output of Llama model
- Using zero3 on multiple nodes is slow
- [BUG]Zero++ training failed
- Docs
- Python not yet supported