I use the gemini-3-pro-image-preview model via AI Studio and the API.
The Issue: * gemini-2.5 text models work fine with my key. * Only gemini-3-pro-prview fails with “The caller does not have permission”.
My Setup & Troubleshooting (Already Done): 1. Billing: My Google Cloud Project is definitely linked to an active Billing Account (Free trial).
2. APIs: I have enabled both the Gemini API in the Cloud Console
Permissions: I am the Project Owner and Billing Admin.
API Key: I created a brand new API key after linking billing. I even tried setting it to “Unrestricted” to rule out scope issues .delete and recreated key again.
It also fails if I add the api key to aistudio chat. “An internal error has occurred.”
Project ID: gen-lang-client-0888209977 Any help would be appreciated as I’ve spent hours debugging this loop.
Yesterday gemini-3-pro-preview was working via the API, now I get a “403 The caller does not have permission” in my script and “Failed to generate content: permission denied. Please try again.” in AI Studio, for Gemini 3.
If there’s a problem, can Google please resolve it, or tell us who to contact for help.
Thanks.
yes, suddenly it start working again on 4 Dec. today (5 Dec ) the api request break down again with different error. It seems like google trying fix it
Hi everyone! So sorry to hear you all have been experiencing errors with the API key. Could you please let me know if the error persists? We are trying to reproduce on our end, however, it looks it’s been fixed. Thank you all!
I guess the “The caller does not have permission” error text transform to the "The model is overloaded. Please try again later.”. even I activate my account instead still the api is unreliable. So the gemini api not ready for use in production on apps? or is there any missing step I missed to get reliable service.