I have been using Gemini 2.5 Pro for coding, and it’s a really good model. However, for the last few days, it has not been responding properly. It stops in the middle of a chat. I’ve been using it through Google Cloud Platform with my Google API key, and I don’t know how to fix this issue. If anyone knows how to help, please let me know.
Having the same problem. extremely high instability of the Gemini 2.5 pro model through the API and only through the API. it keeps outputting the finish reason as STOP but either outputting empty responses or half done responses (stopping mid generation with a STOP reason). It isnt my code or my settings, ive tried it in so many different ways, but it always has a good chance to return the STOP finishreason mid sentence generation. Maybe its the prompt i dont know but i doubt since i use the same exact prompt with the same exact model and settings and it has no problem whatsoever in the Google AI studio.
Same issue here, models like flash seem to work and the issue only happens when using 2.5 pro. Getting a 500 error and a STOP.