I’m currently experiencing consistent crashes with the Antigravity agent and wanted to see if others are facing the same issue or if there’s an ETA for a fix.
I was in the middle of having the agent help me code some server-side logic and port configurations for a Project Zomboid multiplayer server when the agent suddenly terminated. Clicking “Retry” doesn’t seem to bypass the issue for long.
Here is the exact debug log:
Trajectory ID: 707f15f3-1042-4fc6-b24d
Error: HTTP 503 Service Unavailable
Sherlog:
TraceID: 0xe44bb38ed
{
“error”: {
“code”: 503,
“details”: [
{
“@type”: “type.googleapis.com/google.rpc.ErrorInfo”,
“domain”: “cloudcode-pa.googleapis.com”,
“metadata”: {
“model”: “gemini-3.1-pro-high”
},
“reason”: “MODEL_CAPACITY_EXHAUSTED”
}
],
“message”: “No capacity available for model gemini-3.1-pro-high on the server”,
“status”: “UNAVAILABLE”
}
}
I am currently subscribed to the Ultra tier. I understand that server-wide traffic spikes happen, but I wanted to ask the community/team: is there currently a massive global outage for gemini-3.1-pro-high? And how does priority routing work for Ultra subscribers during these capacity-exhausted events?
Any insights or updates from the Google team would be greatly appreciated. Thanks!
I’m experiencing the exact same thing. I spent half the day trying to fix it, thinking it was a problem on my end, before realizing it’s just Google’s servers acting up. I’ve already given up and will just try again later.
Subject: Antigravity IDE 503 MODEL_CAPACITY_EXHAUSTED Ultra Tier
Trajectory ID: 3bd200b5-96e3-426c-93f
TraceID: 0x34cc9a (10:16 MSK)
CloudCode/Antigravity/Code Assist не работает с 11:00 MSK
Ultra tier 25k credits unused, Europe region
Gemini 3.1 Pro Low + Claude Sonnet 4.6 BOTH DOWN
Consumer sites (claude.ai) work fine
Demand: Priority capacity OR full refund