Maybe this is a clue.
I tend to use Claude for heavy tool calling and Gemini for ‘deep’ reasoning so maybe that’s why I didn’t see the Gemini 3 issue with tools until now. It does look like a model glitch that can be prompted around. Thanks
Maybe this is a clue.
I tend to use Claude for heavy tool calling and Gemini for ‘deep’ reasoning so maybe that’s why I didn’t see the Gemini 3 issue with tools until now. It does look like a model glitch that can be prompted around. Thanks