Skip to content

Conversation

@x574chen
Copy link
Collaborator

@x574chen x574chen commented Aug 6, 2025

No description provided.

@x574chen x574chen requested a review from kzjeef August 6, 2025 11:44
Copy link

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

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

Summary of Changes

Hello @x574chen, 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 enhances the flexibility and accuracy of determining the maximum sequence length for models within the system. It shifts from primarily relying on a fixed default or environment variables to dynamically reading the model's inherent maximum position embeddings from its configuration, ensuring that the system's operational limits align more closely with the model's capabilities. This change also provides a significantly increased default maximum length for general use.

Highlights

  • Configuration Update: Increased the default value for the --max_length argument in the API server configuration from 32,000 to 128,000.
  • Dynamic Model Length Determination: Implemented logic to dynamically determine the engine_max_length for the Allspark model configuration. This now considers the max_position_embeddings attribute from the model's configuration (if available) and takes the minimum of that value and the configured --max_length.
  • Runtime Token Limit Alignment: Updated the Qwen-VL runtime to derive its max_total_len directly from the as_config.engine_max_length and max_input_len from max_total_len - 1, removing reliance on DS_LLM_MAX_IN_TOKENS and DS_LLM_MAX_TOKENS environment variables.
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 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 or fill out our survey 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
Copy link

Warning

Gemini is unable to generate a review due to a potential policy violation.

@x574chen
Copy link
Collaborator Author

x574chen commented Aug 6, 2025

Fix #93

@kzjeef kzjeef merged commit 06669fa into modelscope:main Aug 6, 2025
4 checks passed
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