[model-gateway] add qwen2_vl model image processor and tests

[slin1237]

Checklist

• Format your code according to the Format code with pre-commit.
• Add unit tests according to the Run and add unit tests.
• Update documentation according to Write documentations.
• Provide accuracy and speed benchmark results according to Test the accuracy and Benchmark the speed.
• Follow the SGLang code style guidance.
• Work with maintainers to merge your PR. See the PR Merge Process

[gemini-code-assist]

Summary of Changes

Hello @slin1237, 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 significantly enhances the sgl-router's multimodal capabilities by adding support for the Qwen2-VL model. It provides a dedicated image processor that intelligently handles image resizing and preparation, ensuring that visual inputs are correctly formatted for the Qwen2-VL vision encoder. This expansion allows the router to seamlessly integrate and serve a broader range of advanced multimodal models, with the new functionality thoroughly validated against established reference implementations.

Highlights

• New Qwen2-VL Image Processor: Introduced a Rust-native Qwen2VLProcessor to handle the specific dynamic resolution and smart resizing requirements for Qwen2-VL models.
• Smart Resizing Algorithm: Implemented a smart_resize method that preserves image aspect ratios, fits images within configurable pixel bounds, and aligns dimensions to patch boundaries for efficient vision encoding.
• Comprehensive Golden Tests: Added robust golden tests to ensure pixel-perfect compatibility between the Rust Qwen2VLProcessor and the HuggingFace Python implementation across various image sizes and aspect ratios, validating grid dimensions, token counts, and pixel values.
• Integration and Export: The new processor is fully integrated into the sgl-router's image processing registry, making it discoverable and usable, and its module is properly exported.

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 by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

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 pull request 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 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. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

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 Qwen2-VL model, including its image processor and corresponding golden tests. The implementation is comprehensive and well-tested. I've found a couple of areas for improvement in the new qwen2_vl.rs processor implementation. One is a high-priority suggestion to refactor a complex function for better performance and maintainability, and the other is a medium-priority suggestion to remove some redundant code. Overall, great work on adding this new model support.