[Isotr0py]

Purpose

• Actually, --mm-processor-kwargs '{"size": {"longest_edge": 234881024, "shortest_edge": 4096}}' will also update Qwen3-VL image processor's max_pixels.
• So mm_processor_kwargs.get("max_pixels", None) is not robust enough for this case.

Test Plan

Test Result

---

Essential Elements of an Effective PR Description Checklist

• The purpose of the PR, such as "Fix some issue (link existing issues this PR will resolve)".
• The test plan, such as providing test command.
• The test results, such as pasting the results comparison before and after, or e2e results
• (Optional) The necessary documentation update, such as updating supported_models.md and examples for a new model.
• (Optional) Release notes update. If your change is user facing, please update the release notes draft in the Google Doc.

[gemini-code-assist]

Code Review

This pull request correctly updates the logic for calculating the dummy image size for Qwen-VL models. Instead of trying to get max_pixels directly from mm_processor_kwargs, it now correctly initializes the image processor with the provided kwargs and then retrieves the longest_edge from the processor's size attribute. This is a more robust approach. I have one minor suggestion for code simplification in qwen3_vl.py to improve consistency.

[gemini-code-assist]

The or {} here is redundant, as mm_processor_kwargs is guaranteed to be a dictionary by the preceding line (mm_processor_kwargs = mm_processor_kwargs or {}). Removing it will make the code cleaner and more consistent with the changes in the other files in this pull request.

Suggested change

	image_processor = self.info.get_image_processor(**(mm_processor_kwargs or {}))
	image_processor = self.info.get_image_processor(**mm_processor_kwargs)