[Gemini 3.5 Flash] Was temperature and topP removal intentional?

Google AI Studio
, , , ,
1

screenshot
screenshot581×1043 41 KB

I’m not sure if it was intentionally designed this way or not, but could you devs please let us adjust the temperature and topP values again like before?

2

I was wondering the same thing. The Gemma frontier models have the most customization from what I have seen.

3

Yes, this is even stated in their 3.5 flash documentation. Frankly, I think this is a very strange decision, because the larger the context, the lower the top-p and temperature required. 3.5 flash is simply not suitable for long chats because of this nuance…

4

It is intentional, they don’t want you to modify those parameters anymore:

imagen
imagen1042×322 27.3 KB

Depite that, they are still modifiable trough the API.

The screenshot is from https://ai.google.dev/gemini-api/docs/interactions/whats-new-gemini-3.5