mindspore-ai / mindspore-ai/hyper-parallel
[Bug]: 4.57版本的transformers跑llamafactory开启EP后报错
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 53
- Forks
- 63
- Avg merge
- 23h 45m
- Merged PRs (30d)
- 63
Description
Checklist
- 1. I have searched the existing issues (https://gitcode.com/mindspore/hyper-parallel/issues)
- 2. I have read the relevant documentation.
- 3. I have created a minimal reproduction case that clearly demonstrates the issue, including a complete code example and the error message with full traceback and error logs.
🐛 Describe the bug
不同transformers代码许多model shape定义、函数接口不一致,导致不兼容。例如 Qwen3VL模型:
- 专家参数定义不一致


- router的返回参数数量、含义不同
涉及代码:
- hyper_parallel/integration/llamafactory/expert_parallel/models/qwen3_vl_moe.py
- hyper_parallel/integration/llamafactory/utils.py
Expected behavior
兼容4.57.x版本transformers
Additional context
Environment info
transformers 4.57.3
Thanks for contributing 🎉!
schema_version: 1
source: gitcode
gitcode_repo: mindspore/hyper-parallel
gitcode_issue: 380
source_url: https://gitcode.com/mindspore/hyper-parallel/issues/380
Contributor guide
No contributing guide indexed for this repository
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 with hyper_parallel/integration/llamafactory/expert_parallel/models/qwen3_vl_moe.py and hyper_parallel/integration/llamafactory/utils.py, then reproduce the failure with transformers 4.57.3 and LLaMAFactory using expert parallelism. Compare the Qwen3VL expert parameter definitions and router return values with the integration assumptions; done means the Qwen3VL path works with transformers 4.57.x without the reported error.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Active
- Clarity
- Mostly clear
- Newbie friendliness
- 52/100