deepspeedai / deepspeedai/DeepSpeed
[REQUEST] Support Offload deepspeed engine in RLHF training
Open
Nobody has claimed this yet.
enhancement
- Dominant language
- Python
- Stars
- 43.1k
- Forks
- 5k
- Avg merge
- 4d 15h
- Merged PRs (30d)
- 112
Description
See vLLM sleep mode: https://github.com/vllm-project/vllm/pull/11743
This feature is very important in RLHF
such as the hybrid engine mode in OpenRLHF: https://github.com/OpenRLHF/OpenRLHF/commit/c10f02bd650e65f3163216dffca7a21a895463fd
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reading the referenced vLLM sleep mode pull request and the OpenRLHF hybrid engine commit to understand the requested offload behavior in RLHF training. The issue does not name DeepSpeed files or tests; done would require defining and implementing equivalent DeepSpeed engine support and validating it for the described hybrid mode.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- distributed-systems, machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100