remove enforce-eager for llama-3.2-11b-vision

[yyzxw]

part of #44

[gemini-code-assist]

Summary of Changes

Hello @yyzxw, 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 streamlines the runtime configuration for the llama-3.2-11b-vision-instruct model by eliminating the --enforce-eager argument. This change is likely aimed at optimizing the model's execution strategy, potentially leading to improved performance or more efficient resource utilization.

Highlights

• Configuration Update: The --enforce-eager flag has been removed from the customRuntimeArgs within the metadata.yaml file for the llama-3.2-11b-vision-instruct model.

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. ↩

[github-actions]

Model Metadata Validation Results

✅ All validations passed!

Changed files:

• models/meta/llama-3.2-11b-vision-instruct/metadata.yaml

Validation output:

✅ models/meta/llama-3.2-11b-vision-instruct/metadata.yaml - Valid


✅ models/meta/llama-3.2-11b-vision-instruct/metadata.yaml - Valid directory structure


✅ models/meta/llama-3.2-11b-vision-instruct/metadata.yaml - Valid Hugging Face source: meta-llama/Llama-3.2-11B-Vision-Instruct
⚠️ models/meta/llama-3.2-11b-vision-instruct/metadata.yaml - Organization names differ: HF:meta-llama vs MS:llm-research
✅ models/meta/llama-3.2-11b-vision-instruct/metadata.yaml - Valid ModelScope source: LLM-Research/Llama-3.2-11B-Vision-Instruct


⚠️ models/meta/llama-3.2-11b-vision-instruct/metadata.yaml - Hugging Face name (meta-llama/Llama-3.2-11B-Vision-Instruct) and ModelScope name (LLM-Research/Llama-3.2-11B-Vision-Instruct) don't match
⚠️ models/meta/llama-3.2-11b-vision-instruct/metadata.yaml - Metadata name (llama-3.2-11b-vision-instruct) doesn't match the normalized Hugging Face model ID (llama-3-2-11b-vision-instruct)

Please make sure all metadata files conform to the schema defined in docs/model.jsonschema.

[gemini-code-assist]

Code Review

This pull request removes the --enforce-eager flag for the llama-3.2-11b-vision-instruct model. This change is likely intended to improve performance by enabling optimizations like CUDA graphs in vLLM. However, the pull request lacks a description or code comments explaining why this change is safe and what validation has been performed. I have added a comment highlighting the potential correctness risk and requesting that this context be added to improve maintainability and allow for a proper review.

[yyzxw]

有一点需要注意的是： vllm 最新版中启动 llama-3.2 会失败：https://github.com/vllm-project/vllm/issues/27198， 使用 VLLM_USE_V1=0 去启动也会失败