[Misc][Main2Main] Upgrade vLLM to v0.20.1 and 0506

[wxsIcey]

What this PR does / why we need it?

Fixes NPUInputBatch not have thinking_budget_state_holder = None, caused by [Reasoning][Feature] Support for speculative decoding with thinking budget

Fixes 'AscendMultiHeadLatentAttention' not have skip_topk, caused by [Feature]: IndexCache support for DSA models

Fixes MLA prefill backends selection, caused by [Attention] Abstract the MLA prefill backends and eliminate cuDNN

Fixes ModelRunner V2 eagle refactor, caused by [Model Runner V2] Skip attention metadata rebuild before draft prefill
, [Model Runner V2] Rebuild attn metadata between draft decode steps
, [Model Runner V2] Add logprob_token_ids support, [Model Runner V2] Fix rejection sampling acceptance rate gap vs MRV1

Does this PR introduce any user-facing change?

N/A

How was this patch tested?

CI passed with new added/existing test.

• vLLM version: v0.19.1
• vLLM main: vllm-project/vllm@4d51588

[github-actions]

👋 Hi! Thank you for contributing to the vLLM Ascend project. The following points will speed up your PR merge:‌‌

• A PR should do only one thing, smaller PRs enable faster reviews.
• Every PR should include unit tests and end-to-end tests ‌to ensure it works and is not broken by other future PRs.
• Write the commit message by fulfilling the PR description to help reviewer and future developers understand.

If CI fails, you can run linting and testing checks locally according Contributing and Testing.

[gemini-code-assist]

Summary of Changes

Hello, 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 upgrades the vLLM dependency to version 0.20.1 and introduces necessary adjustments to the Ascend NPU integration. The changes ensure compatibility with upstream vLLM refactors, specifically addressing missing attributes in input batches and attention layers, and providing robust handling for MLA prefill backends and speculative decoding workflows.

Highlights

• Dependency Upgrade: Upgraded the vLLM dependency to version 0.20.1 across all Dockerfiles and documentation.
• Compatibility Fixes: Added missing attributes to NPUInputBatch and AscendMultiHeadLatentAttention to maintain compatibility with upstream vLLM changes.
• MLA Prefill Backend: Introduced a new patch for the MLA prefill backend to prevent runtime crashes on Ascend NPUs by registering a no-op implementation.
• ModelRunner V2 Refactor: Updated Eagle speculative decoding and logprob sampling logic to support version-conditional execution for v0.20.1.

New Features

🧠 You can now enable Memory (public preview) 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.

Ignored Files

• Ignored by pattern: .github/workflows/** (7)
  • .github/workflows/_e2e_test.yaml
  • .github/workflows/dockerfiles/Dockerfile.lint
  • .github/workflows/pr_test_full.yaml
  • .github/workflows/pr_test_light.yaml
  • .github/workflows/schedule_update_estimated_time.yaml
  • .github/workflows/schedule_vllm_e2e_test.yaml
  • .github/workflows/scripts/config.yaml

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 the 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 counterproductive. 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.

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 upgrades the vLLM dependency to v0.20.1 and implements compatibility fixes for the Ascend NPU backend, covering MLA prefill handling, input batch structures, and Eagle speculative decoding. The reviewer identified a critical recurring logic error where version checks were inverted, which would disable compatibility logic on the target version. A redundant variable assignment in the MLA module was also noted.\n\nSuggested PR Title:\n\nmarkdown\n[Ops][Misc] Upgrade vLLM to v0.20.1 and compatibility fixes\n\n\nSuggested PR Summary:\n\nmarkdown\n### What this PR does / why we need it?\nThis PR upgrades the vLLM dependency to v0.20.1 and includes several compatibility fixes for the Ascend NPU backend:\n- Fixes `NPUInputBatch` compatibility with thinking budget state.\n- Fixes `AscendMultiHeadLatentAttention` missing `skip_topk` attribute.\n- Fixes MLA prefill backend selection to avoid crashes on Ascend.\n- Updates ModelRunner V2 Eagle speculative decoding to match vLLM 0.20.1 refactors.\n\n### Does this PR introduce _any_ user-facing change?\nNo.\n\n### How was this patch tested?\nCI passed with existing tests.\n

[github-actions]

This pull request has conflicts, please resolve those before we can evaluate the pull request.

[wxsIcey]

if we do not want test model runner v2 on 0.20.1, we need delete it. Otherwise many import problems will occur.

[wxsIcey]

Please see this workflow: https://github.com/vllm-project/vllm-ascend/actions/runs/25605210483/job/75165950902?pr=8983