Problem with reasoning_effort Parameter When Using O4-mini Model

[lam584]

Software Version

cherry-studio 1.5.3 (Windows application)

Description

When calling OpenAI's O4-mini API model, I tried to change the reasoning_effort parameter from the default "medium" to "high", but I received the following error:

400 Unsupported parameter: 'reasoning_effort'. In the Responses API, this parameter has moved to 'reasoning.effort'. Try again with the new parameter. See the API documentation for more information: https://platform.openai.com/docs/api-reference/responses/create.

According to OpenAI’s official documentation, it is recommended to use reasoning.effort as the parameter.

However, when I change the parameter name to reasoning.effort, I get a different error:

400 Unknown parameter: 'reasoning.effort'.

Relevant documentation: https://platform.openai.com/docs/api-reference/responses/create

---

Main Question
What is the correct way to configure this parameter for the O4-mini model?

---

Steps to Reproduce

1. Use the cherry-studio application to call OpenAI's O4-mini model API.
2. Try setting the reasoning_effort parameter to "high" (error as above).
3. Try setting the parameter name to reasoning.effort (error as above).

---

Screenshots and Error Messages

---

Any help would be greatly appreciated. Thank you!

[DeJeune]

remove custom parameter and choose a reasoning option in InputBar:

[lam584]

Thank you

[sleepyy-dog]

remove custom parameter and choose a reasoning option in InputBar:删除自定义参数并在 InputBar 中选择一个推理选项：

i want to ask if when i use gpt5 with reasoning_effort as 'high',this solution can also solve my requirement?

when i set my custom parameter,the problem is below,two parameter is conflicted：