Adjusting LLM parameters doesn't seem to pass relevant flags to the cloud provider. I checked OpenAI and Google (in AI Studio) logs, and reasoning effort parameters don't show up in the request.
Are you on the latest version (v2.3.0)? I can’t seem to be able to reproduce the issue with OpenAI.
For Gemini, it’s expected at the moment. As Google AI doesn’t support the “reasoning effort” parameter. I’ll add support for their thinkingBudget and thinkingLevel params soon.
Aaron Kunaseelan
•
Dec 4, 2025
Yes i’m on (v2.3.0); i tried with and without tool use (without tool use example below)
Daniel Nguyen
Team•
Dec 4, 2025
Thank you. I assume you use the official OpenAI provider and not via a proxy, or using a custom OpenAI-compatible provider. It’s strange that I couldn’t reproduce the issue. Do you have any custom configuration on the provider?
Log in to comment and vote
Comments9
Aaron Kunaseelan
Dec 11, 2025
Do you have any new info on the issue?
Aaron Kunaseelan
Dec 11, 2025
@Daniel Nguyen seems fixed on update to tahoe
Daniel Nguyen
Dec 12, 2025
Glad to hear that. Thanks
Daniel Nguyen
Dec 4, 2025
Are you on the latest version (v2.3.0)? I can’t seem to be able to reproduce the issue with OpenAI.
For Gemini, it’s expected at the moment. As Google AI doesn’t support the “reasoning effort” parameter. I’ll add support for their thinkingBudget and thinkingLevel params soon.
Aaron Kunaseelan
Dec 4, 2025
Yes i’m on (v2.3.0); i tried with and without tool use (without tool use example below)
Daniel Nguyen
Dec 4, 2025
Thank you. I assume you use the official OpenAI provider and not via a proxy, or using a custom OpenAI-compatible provider. It’s strange that I couldn’t reproduce the issue. Do you have any custom configuration on the provider?
I’m trying my best to reproduce it.