Enhancement: Add setting to hide "thought" responses from Google Gemini Thinking Mode models

[robinvalk]

What features would you like to see added?

Currently the Google Gemini Thinking Mode model responses come in "two parts". As from the Google docs:

On a standard request, the model responds with two parts, the thoughts and the model response.

It looks like that in the current implementation both of the parts are displayed in the chat. It would be nice to have a setting to disable the "thought" response messages to be displayed.

More details

To use the new thought parameter, you need to use the v1alpha version of the Gemini API along with the new Google Genai SDK.

https://ai.google.dev/gemini-api/docs/thinking-mode#working_with_thoughts

Which components are impacted by your request?

No response

Pictures

Picture of the very expressive responses

Code of Conduct

• I agree to follow this project's Code of Conduct

[voelspriet]

Just use gemini-2.0-flash-exp as model

[tedlin0124]

thinking model is different with default model. The output is better.

[michnovka]

This would be a good thing for more models, perhaps this switch could be more general? Deepseek also has the thought output.

If nothing else, it could be collapsed by default and on click it would expand to show what it was thinking about