modelscope / modelscope/ms-swift
Is it possible to enable sequence parallel with custom model?
Open
Nobody has claimed this yet.
stale
- Dominant language
- Python
- Stars
- 15.7k
- Forks
- 1.7k
- Avg merge
- 1d 16h
- Merged PRs (30d)
- 136
Description
I want to modify embeddings and add some adapters to Qwen3 model(no modification to the transformer layers), so I have to create a custom model. Is it possible to enable sequence parallel with the custom model?
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
No file, test, or entry point is named. Start by locating the custom-model integration and the sequence-parallel implementation, then determine whether the custom Qwen3 model can use that path without transformer-layer changes; done would require a documented, working answer or supported implementation.
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
- 20/100