modelscope / modelscope/ms-swift

GLM4.1V Video Inference Bug Inquiry

Open
#6,528 5 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement stale
Dominant language
Python
Stars
15.7k
Forks
1.7k
Avg merge
1d 16h
Merged PRs (30d)
136

Description

Hi, @hjh0119
Thanks for your pr.
I want to train 4.1v using videos, but there are todos in code. How to solve these bugs? Is there any other ways to train 4.1v in videos?

https://github.com/modelscope/ms-swift/blob/bf0257370abf5eddab7814fef166d87d0e25d557/swift/llm/template/template/glm.py#L133C3-L133C38

https://github.com/modelscope/ms-swift/blob/bf0257370abf5eddab7814fef166d87d0e25d557/swift/llm/template/template/glm.py#L171

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reading the TODOs at lines 133 and 171 of swift/llm/template/template/glm.py, then trace how GLM4.1V video inputs reach this template. The issue does not define expected behavior or a testable fix; clarify the intended training path and reproduce the reported video-training problem before implementation.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
computer-vision, machine-learning
Issue type
Bug
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
18/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.