modelscope / modelscope/ms-swift
qwen3 embedding 训练后模型参数文档变动
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 15.7k
- Forks
- 1.7k
- Avg merge
- 1d 16h
- Merged PRs (30d)
- 136
Description
model.safetensors.index.json config.json 都发生了变化，而且模型的safetensor文件大小也与原模型不同，造成了其他部署方式报错
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 comparing the original and post-training model.safetensors.index.json, config.json, and safetensor file sizes shown in the report. Reproduce the Qwen3 embedding training output and the deployment failure, then trace the model-saving path responsible for the changed artifacts. Done means the trained model preserves the expected parameter metadata and can be deployed through the affected deployment methods.
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
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100