While generating response, the agent (I tried only gemini models so far) suddenly goes into this loop of “thinking, generating, loading” even on simple prompts. I had to stop it and prompt “continue” and it immediately resumes correctly. So there is not heavy thinking going on. It’s probably a bug. Happens in both “fast” and “planning” modes.
Agreed. Up until a day and a half (ish)ago they were hustling plans, fixing code, and minding a couple of other Sparks off on sub-quests for him, and not a facet of the GEM that’s driving him
Thank you for bringing this to our attention. We have escalated the issue to our internal teams for a thorough investigation.
To ensure our engineering team can investigate and resolve these issues effectively, we highly recommend filing bug reports directly through the Antigravity in-app feedback tool. You can do this by navigating to the top-right corner of the interface, clicking the Feedback icon, and selecting Report Issue.
The issue is still happening. It is able to think and make a markdown plans but it is not able to code or do anything it just stucks in an infinite loop every single time no matter what model I change to
It’s sad for me as a former Googler myself to see such a promising product being swarmed by so many issues that I had to completely abandon it and go use Cursor, Codex, or even back to VS code.
I really enjoyed it at the beginning but the numerous bugs and changing rate limits has made Antigravity unusable for me. 
It happens today using 3.5 flash