fix: Change the max_token of the Qianfan large model to max_output_token

[shaohuzhang1]

fix: Change the max_token of the Qianfan large model to max_output_token

[f2c-ci-robot]

Adding the "do-not-merge/release-note-label-needed" label because no release-note block was detected, please follow our release note process to remove it.

Details

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[f2c-ci-robot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:

The full list of commands accepted by this bot can be found here.

Details

Needs approval from an approver in each of these files:

• OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[shaohuzhang1]

The proposed change from max_tokens to max_output_tokens is valid and aligns with the current naming conventions in many APIs, which often use "output" at the end of parameter names to indicate they refer to what an API generates rather than receiving. This makes sense given the context of LLM (Language Model) parameters, especially since models typically output tokens or sentences.

Potential Improvements:

1. Documentation Clarity: Consider adding more detailed documentation for both fields (max_output_tokens and potentially _step, precision) about their intended usage and expected values. This will help users better understand how these parameters affect the behavior of your application.

2. Validation: Ensure that the validation logic remains appropriate for both fields:

   • For max_output_tokens, set a reasonable upper limit based on typical text generation tasks.
   • Validate _step to ensure it meets expectations (e.g., non-negative numbers).
   • Validate precision similarly, ensuring it's within acceptable bounds.

3. Consistent Naming: If there are other slider fields involving token limits (like max_input_tokens), ensure consistent naming across all similar fields to maintain clarity and ease of use.

Overall, this refactoring appears to be well-intentioned and should work correctly as long as the existing codebase and business requirements are adhered to.