Paid for AI Pro.
The quota for all Gemini 3.1 and Claude 4.6 models was above 40%.
My prompt was 429’d due to high peak demand. I waited >1min as requested.
The “continue” was also 429’d.
All the Claude 4.6 quotas were instantly reset to 0% and and 5 days wait.
All the Gemini 3.1 quotas instantly dropped to around 20%.
Honest question: @Abhijit_Pramanik is this a bug, or punishment for retrying during peak times?
Does anyone know when these outages will end? Our work has been disrupted and we are in a very difficult situation.
All quotas for all models were 100% today.
A prompt was 429’d, so I waited ~2.5mins.
The “continue” was also 429’d.
Did not resend. Decided to wait it out for 20mins.
10% of the quota simply disappeared about 15mins later, with no messages actual sent or work done since the first 429.
Can confirm - after receiving the message about high demand, the next “continue” causes ~10% quota to be deducted, even if the “continue” was also 429’d.
In other words:
“Error Our servers are experiencing high traffic right now, please try again in a minute.”
… wait 2 mins then “continue”…
“Error Our servers are experiencing high traffic right now, please try again in a minute.”
…do nothing for a few mins…
~10% deducted from quota.
Note that zero work has been carried out and the quota is reduced.
Version 1.19.6
Model claude-opus-4-6-thinking
Thank you for bringing these concerns to our attention. Please be assured that I have shared your feedback with our internal team for further review.
We appreciate your continued patience as we work to enhance the Antigravity experience.
The quota for the Claude models was exhausted, and the Model Quota screen showed that it would refresh in 4hrs 59mins.
I switched to Gemini 3.1 Pro and kept an eye on the quotas. On receiving the first 429 notification with Gemini, the 3.1 Pro quota dropped by ~20% and then the Claude quotas dropped to zero with a ~7 day refresh.
Note: I’m only using the Claude models because Gemini has been really quirky lately.
