modelscope / modelscope/ms-swift
是混合多模态训练，换模型可以跑起来
Open
Nobody has claimed this yet.
bug
stale
- Dominant language
- Python
- Stars
- 15.7k
- Forks
- 1.7k
- Avg merge
- 1d 16h
- Merged PRs (30d)
- 136
Description
训练脚本发一下，是不是混合模态数据训练
Originally posted by @Jintao-Huang in #6811
是的是的，我混合多模态数据训练
但是我把minicpm -v 2.6换成qwen3_vl就可以正常跑了
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 reproducing the mixed multimodal training case described in the issue, comparing MiniCPM-V 2.6 with Qwen3-VL. The issue does not name a script, error, or test, so first identify the training entry point and capture the failure; done means explaining or correcting the model-specific behavior.
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