Skip to content

Fix MiniMax (docs and integration tests checkpoint) #38575

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Jun 6, 2025

Conversation

geetu040
Copy link
Contributor

@geetu040 geetu040 commented Jun 4, 2025

What does this PR do?

This PR updates MiniMax from #35831 with minor changes

Before submitting

  • This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).

Who can review?

@LysandreJik, since you merged the original PR, just addressing a couple of items. Thanks and sorry for a late follow-up!

CC: @ArthurZucker, @stevhliu

@geetu040 geetu040 force-pushed the minimax-minor-fixes branch from 37644b1 to 4647a7c Compare June 5, 2025 07:06
@stevhliu stevhliu force-pushed the minimax-minor-fixes branch from 4647a7c to 7c2593e Compare June 5, 2025 18:21
Copy link
Member

@stevhliu stevhliu left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for the fixes!

@ydshieh
Copy link
Collaborator

ydshieh commented Jun 5, 2025

Hi @geetu040 Could you share the reason we use such tiny checkpoint (with random weights right?) for integration tests? Is it because otherwise we get GPU OOM?

@geetu040
Copy link
Contributor Author

geetu040 commented Jun 5, 2025

Hi @geetu040 Could you share the reason we use such tiny checkpoint (with random weights right?) for integration tests? Is it because otherwise we get GPU OOM?

yes @ydshieh, the original model, MiniMax-Text-01, has 456B parameters.

@ydshieh
Copy link
Collaborator

ydshieh commented Jun 5, 2025

run-slow: minimax

Copy link
Contributor

github-actions bot commented Jun 5, 2025

This comment contains run-slow, running the specified jobs:

models: ['models/minimax']
quantizations: [] ...

@ydshieh
Copy link
Collaborator

ydshieh commented Jun 5, 2025

I will merge tomorrow once pytest/custom-tests is green. If it's failing just because the expected values and actual values not matching, I will update.

@ydshieh ydshieh merged commit 31023b6 into huggingface:main Jun 6, 2025
13 of 17 checks passed
bvantuan pushed a commit to bvantuan/transformers that referenced this pull request Jun 12, 2025
)

* update checkpoints for integration tests

* minor fixes in docs
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants