We are having a quite annoying issue with Gemini 2.5 Flash thinking mode, it overshoots the thinking budget by A LOT, eating pretty much the whole output length. If happens like this: We set the thinkingBudget to a value like 1512 and the maxOutputTokens to 4000, most of the time it keeps between 1…

Gemini 2.5 Flash Overthinking by a lot

cor September 5, 2025, 2:45pm 13

I am noticing the same issue and just made a post and shared my steps here

1 Like