Add one-step recipe serving by louie-tsai · Pull Request #57306 · vllm-project/vllm · GitHub
Skip to content

Add one-step recipe serving - #57306

Merged
bigPYJ1151 merged 1 commit into
vllm-project:mainfrom
intel-ai-tce:recipes-one-step-serving
Sep 20, 2026
Merged

bigPYJ1151 merged 1 commit into
vllm-project:mainfrom
intel-ai-tce:recipes-one-step-serving

Conversation

@louie-tsai

@louie-tsai louie-tsai commented Sep 17, 2026 •

Copy link
Copy Markdown
Contributor

Purpose

Add a simple entry point that generates a configuration from vLLM Recipes and starts the vLLM server in one step.

Example:

tools/recipes/serve_with_recipe.sh \
  --model meta-llama/Llama-3.1-8B-Instruct \
  --hardware xeon6

The script:

  • generates config.yml and env.sh
  • enables hardware detection automatically for Xeon 6
  • loads the generated environment
  • starts from /vllm-workspace when recipe-relative assets are stored there
  • preserves generated output files in the caller's working directory
  • launches vllm serve with the generated configuration

This PR is split from #56340 and is independent of the staged sweep workflow.

Test Plan

bash -n tools/recipes/serve_with_recipe.sh

tools/recipes/serve_with_recipe.sh \
  --model meta-llama/Llama-3.1-8B-Instruct \
  --hardware xeon6

Test Result

tested locally


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.

@claude claude Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Claude Code Review

This pull request is from a fork — automated review is disabled. A repository maintainer can comment @claude review to run a one-time review.

@louie-tsai louie-tsai mentioned this pull request Sep 17, 2026
4 tasks
@louie-tsai
louie-tsai force-pushed the recipes-one-step-serving branch from 4eaefd8 to 7d3db89 Compare September 17, 2026 06:59
@louie-tsai

Copy link
Copy Markdown
Contributor Author

Signed-off-by: louie-tsai <louie.tsai@intel.com>
@louie-tsai
louie-tsai force-pushed the recipes-one-step-serving branch from 7d3db89 to 52ab98f Compare September 18, 2026 05:41
@bigPYJ1151

Copy link
Copy Markdown
Member

/ci run

1 similar comment
@bigPYJ1151

Copy link
Copy Markdown
Member

/ci run

@github-actions

Copy link
Copy Markdown

❌ This PR is 6 commits behind upstream main. Your branch must contain every commit currently on upstream main. No new CI build was started. Merge or rebase onto the latest main, then rerun /ci run. To test this branch at your own risk, use /ci run --allow-stale.

@bigPYJ1151

Copy link
Copy Markdown
Member

/ci run --allow-stale

@github-actions

Copy link
Copy Markdown

@bigPYJ1151
bigPYJ1151 merged commit 03f8301 into vllm-project:main Sep 20, 2026
20 checks passed
michael-gpu pushed a commit to practical-inference/vllm-ampere-extended that referenced this pull request Oct 20, 2026
Signed-off-by: louie-tsai <louie.tsai@intel.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants