deepspeedai / deepspeedai/DeepSpeed
[BUG] Deepspeed inference engine seems to only support tensor parallelism and not model parallelism; Can we partition a model into pipeline stages via the inference engine support in deep speed ?
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 43.1k
- Forks
- 5k
- Avg merge
- 4d 15h
- Merged PRs (30d)
- 112
Description
Describe the bug
(https://www.deepspeed.ai/tutorials/inference-tutorial/) seems to only support tensor parallelism and not model parallelism;
Can we partition a model into pipeline stages via the inference engine support in deep speed ?
Any dev repo which can be used for pipeline stages for inferencing via deep speed.
One of the texts mentions below:
DeepSpeed Inference currently supports tensor slicing–based multi-GPU parallelism for Transformer-based models, both within and across nodes, and we plan to add support for pipeline parallelism soon.
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
Begin with the linked inference tutorial and compare its documented tensor-slicing support with the requested pipeline-stage behavior. The issue does not identify implementation files, tests, or a defined acceptance criterion; a contributor would need maintainers to point to the inference entry point and clarify whether pipeline parallelism is in scope.
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
- 15/100