Feature: Add More LLM Request Parameter Settings

[wojiaoyishang]

Self Checks

• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
• [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:）
• Please do not modify this template :) and fill in all the required fields.

1. Is this request related to a challenge you're experiencing? Tell me about your story.

Different language models have various adjustable request parameters. Although the LLM generation node comes pre-configured with parameters such as Top-P and temperature, there are additional parameters that vary from different vendors to individual models. For example, the OpenAI documentation also provides adjustments for parameters like reasoning_effort and web_search_options.

In third-party interfaces (such as the qwen2.5-vl model), there is a vl_high_resolution_images parameter to control whether images need to be resized.

2. Additional context or comments

Here, I sincerely propose this feature, hoping that custom parameter settings can be added in the next version of dify. Since this is aimed at advanced users, a simple addition could allow users to pass a JSON-formatted data to include extra parameters.

Thanks!

3. Can you help us with this feature?

• I am interested in contributing to this feature.