Remove unused swap_space parameter

[mcelrath]

References: #27984

Purpose

Remove unused swap_space parameter after the removal of best_of sampling. Fixes #27984

Test Plan

(Untested) working on it.

Help would be appreciated. I'm having quite a lot of difficulty compiling vllm and all its dependencies. This should be a simple fix with no impact anywhere else.

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.

[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 ask your reviewers to trigger select CI tests on top of fastcheck CI.

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.

If you have any questions, please reach out to us on Slack at https://slack.vllm.ai.

🚀

[mergify]

Documentation preview: https://vllm--27988.org.readthedocs.build/en/27988/

[gemini-code-assist]

Code Review

This pull request aims to remove the unused swap_space parameter throughout the codebase. The changes are comprehensive, touching documentation, tests, and core configuration files. While most of the changes are correct, I've identified a critical issue in vllm/engine/arg_utils.py that appears to be a typo. This typo will likely cause a KeyError during command-line argument parsing, breaking the CLI. I've provided a suggestion to fix this.

[gemini-code-assist]

There seems to be a typo here. The get_kwargs(CacheConfig) function returns a dictionary with keys corresponding to the fields in CacheConfig. CacheConfig has a field named cache_dtype, but not kv_cache_dtype. Accessing cache_kwargs["kv_cache_dtype"] will likely cause a KeyError. This should probably be cache_kwargs["cache_dtype"] as it was before this change.

Suggested change

	cache_group.add_argument("--kv-cache-dtype", **cache_kwargs["kv_cache_dtype"])
	cache_group.add_argument("--kv-cache-dtype", **cache_kwargs["cache_dtype"])

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Fix CLI lookup for kv cache dtype option

The new code builds the CLI by fetching cache_kwargs["kv_cache_dtype"], but the CacheConfig dataclass still exposes the field as cache_dtype. get_kwargs(CacheConfig) therefore produces a dictionary with key "cache_dtype", and this lookup now raises a KeyError as soon as the argument parser is constructed (e.g. running python -m vllm.entrypoints.llm --help). This prevents the CLI from starting at all. Keep the lookup keyed on "cache_dtype" or rename the field consistently.

Useful? React with 👍 / 👎.

[hmellor]

I see a few more references that aren't removed by this PR:

cc @ApostaC does LMCache depend on the swap_space argument at all in vllm/distributed/kv_transfer/kv_connector/v1/lmcache_integration/vllm_v1_adapter.py?

[ApostaC]

Hey @hmellor , we don't use that at all.

[DarkLight1337]

cc @njhill

[mergify]

Hi @mcelrath, the pre-commit checks have failed. Please run:

uv pip install pre-commit
pre-commit install
pre-commit run --all-files

Then, commit the changes and push to your branch.

For future commits, pre-commit will run automatically on changed files before each commit.

[github-actions]

This pull request has been automatically marked as stale because it has not had any activity within 90 days. It will be automatically closed if no further activity occurs within 30 days. Leave a comment if you feel this pull request should remain open. Thank you!