[model-gateway] Fix tokenizer caching and improve error handling

[slin1237]

• Cache tokenizer in RequestContext to avoid redundant registry lookups (4 lookups per request -> 1 lookup, reused across pipeline stages)
• Replace dangerous .unwrap() on model_id with proper error handling
• Add UNKNOWN_MODEL_ID constant to eliminate magic strings
• Fix copy-paste errors in error log function names
• Remove unused tokenizer() helper method (dead code)

Files changed:

• context.rs: Add tokenizer field to ProcessingState, add tokenizer_arc() helper
• chat/preparation.rs, generate/preparation.rs: Cache tokenizer, fix model_id unwrap
• chat/response_processing.rs, generate/response_processing.rs: Use cached tokenizer
• core/mod.rs: Add UNKNOWN_MODEL_ID constant
• model_card.rs, dispatch_metadata.rs, bucket.rs, cache_aware.rs: Use constant

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 model-gateway's performance and reliability. By implementing tokenizer caching, it optimizes the request processing pipeline, reducing overhead. Concurrently, it bolsters error handling mechanisms, particularly for model identification, preventing runtime panics and providing clearer error messages. The introduction of a dedicated constant for unknown model IDs further refines code quality and consistency across the system.

Highlights

• Tokenizer Caching: The tokenizer is now cached within the RequestContext's ProcessingState, reducing redundant registry lookups from 4 to 1 per request across pipeline stages, improving performance.
• Improved Error Handling: Potentially crashing .unwrap() calls on model_id have been replaced with robust error handling, returning proper internal errors when model_id is not set, enhancing system stability.
• Constant for Unknown Model ID: A new UNKNOWN_MODEL_ID constant has been introduced to eliminate magic string literals, improving code maintainability and consistency across various modules.
• Error Log Function Name Fixes: Copy-paste errors in error log function names within the generate preparation and response processing stages have been corrected for more accurate debugging.
• Dead Code Removal: An unused tokenizer() helper method has been removed, contributing to a cleaner and more concise codebase.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

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 introduces several valuable improvements. Caching the tokenizer in the RequestContext is a great performance optimization that reduces redundant lookups. Replacing .unwrap() calls with proper error handling significantly improves the robustness of the code. Furthermore, introducing the UNKNOWN_MODEL_ID constant and fixing incorrect function names in logs enhance code quality and maintainability. The changes are well-implemented and clearly beneficial. I have one suggestion to further improve maintainability by reducing some code duplication.