[Spec Decode] Refactor spec decoding into a separate function

[WoosukKwon]

Currently, the spec decoding logic is embedded in the execute_model method, making it long and a bit difficult to read. This PR carves out the spec decoding code and moves it into the separate method propose_draft_token_ids.

[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 @WoosukKwon, 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 focuses on improving the modularity and readability of the gpu_model_runner.py file. It achieves this by extracting the complex speculative decoding logic from the main execute_model method into a new, dedicated helper function, propose_draft_token_ids. This change streamlines the primary execution flow, making the code easier to follow and maintain, without altering the functional behavior of the speculative decoding process.

Highlights

• Code Refactoring: The core speculative decoding logic has been extracted from the lengthy execute_model method into a new, dedicated method named propose_draft_token_ids.
• Improved Readability: This refactoring significantly reduces the complexity and length of execute_model, making it easier to understand and maintain by isolating a specific functional block.
• Parameter and Method Renames: Updated parameter names (e.g., valid_sampled_token_ids to sampled_token_ids) and renamed the helper method generate_draft_token_ids to propose_ngram_draft_token_ids for consistency within the new function's context.
• Minor Logic Adjustments: Included explicit initialization for aux_hidden_states and updated how positions are accessed (from positions to self.positions) within the moved speculative decoding logic.

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 refactors the speculative decoding logic into a new propose_draft_token_ids method, which improves code clarity. M-RoPE support is broken due to the positions tensor not being passed to the new function, and I've included suggestions to address this.