[Model] Support HF format of minimax

[mgoin]

Purpose

#20199 set us up to test the older format of MiniMaxAI/MiniMax-Text-01 we initially implemented. This PR implements the small changes to support the new HF format within the existing model definition and the new explicit format at https://huggingface.co/MiniMaxAI/MiniMax-Text-01-hf . This was based on looking at the changes to the model config and basically amounts to renames for existing configs https://huggingface.co/MiniMaxAI/MiniMax-Text-01/commit/130e821701f3244ab18bab619dbd6fa9b5a92d28

Test Plan

Test using the registry. It would be nice to run an evaluation, but I don't have a machine to load the ~460B model atm.

Test Result

[github-actions]

👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can either: Add ready label to the PR or enable auto-merge.

🚀

[gemini-code-assist]

Summary of Changes

Hello @mgoin, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request enables vLLM to support the latest Hugging Face configuration format for the MiniMaxAI/MiniMax-Text-01 model. It involves adapting the model's parameter parsing logic to accommodate new naming conventions and structures for attention and normalization layers, ensuring that models saved in this new format can be loaded and run correctly.

Highlights

• New Model Format Support: Added support for the new Hugging Face configuration format for the MiniMaxAI/MiniMax-Text-01 model, allowing vLLM to correctly interpret its updated structure.
• Configuration Parsing Updates: Modified the minimax_text_01 model's initialization to dynamically retrieve layernorm alpha/beta factors and attention layer types (linear_attention vs. full_attention) from the new HF config's naming conventions, including a fallback mechanism for parameters and a mapping for layer types.
• Internal Consistency: Updated internal references within the minimax_text_01 model from self.config.attn_type_list to self.model.decoder_attention_types to align with the newly parsed attention type information, ensuring correct calculation of flash layer counts and weight loading.
• Registry Updates: Registered the new MiniMaxForCausalLM model name in both the test and main model registries, linking it to the existing minimax_text_01 implementation.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point in your pull request via creating an issue comment (i.e. comment on the pull request page) using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in issue comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist is currently in preview and may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments to provide feedback.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request adds support for the new Hugging Face format of MiniMaxAI/MiniMax-Text-01 while maintaining backward compatibility with the older format. The changes are logical and correctly implemented. I've provided a couple of suggestions to improve code readability and maintainability by using more idiomatic Python constructs.

[ywang96]

FYI @qscqesze

[qscqesze]

Sorry for the late response — I just saw this PR. It's great, LGTM!

[qscqesze]

The -hf repository isn't our main focus; it's a special version tailored for the transformers library, created because some config parameters conflicted with the original framework. In vLLM, it's optional and unlikely to be used for deploying a version meant specifically for transformers.

[mgoin]

Okay, thanks a lot for the context @qscqesze. I only made this because the original model checkpoint was changed on this commit but I see now it has been changed back.

I'll still work on adding the HF support since new quantized versions could use it.