Something is definitely wrong with the Model in the past couple of weeks

I’m a VP of Eng and have been developing both in personal projects and work projects with AG for the past 2 month and have enjoyed every moment (mostly using Gemini Flash 3 variations). The speed of development was so great and the quotas hit the right spot.

About 1-2 weeks ago AG start to become completely useless, I can’t get it to do even basic things in the projects I’ve worked on, even after restructuring the files for better context window optimization.

Testing the claude opus 4.5 response time: seems that Cursor can provide the same response at exactly half the speed.
While I was about to switch my company from Cursor to AG, I don’t think I’ll proceed if this persist.

@Google, I do hope you’re listening, acknowledge and react…

Hello @AG_Coder, welcome to AI Forum!

Thank you for sharing your candid feedback. We are actively working on improvements to ensure the model performance meets the high standards you expect. We sincerely appreciate your patience and support as we continue to refine the experience.

1 Like