“First time?” (c)
Wait a couple more weeks, and the support rep will probably show up and say, “We discovered an issue and we’re working on it”. Google moment. SSDD.
the problem isnt that bad with claude models, but google models? HELLA unusable.
like, for every prompt, i need to retry atleast 3 times. aka, about twice a minute.
the only reason i’d been using antigravity at all is because its quite fine at web dev and the google models have a pretty okay ratelimit, but now those models are just straightoff unusable because of these interruptions
whats worse is that it doesnt hesitate to straightaway instantly cut any edit in between, so you gotta retry it again, and then the model completely steers away.
I just cancled. Antigravity has become unusable at this point. As long as there is not even a proper reaction to their degraded service I cant take this anymore.
Hello everyone!
Thank you for bringing the issue to our attention. Our engineering team is currently investigating the matter, and we appreciate your patience as we work toward a resolution.
Yep, I have Claud working but Flash isnt working at all, I am still stuck in the 6 day purgatory for Pro class, with only Opus and Sonnet working.
I appreciate that this was in BETA when I signed up, but as someone who had a previous life in Ops, and now runs enterprise innovation, I have to say this support is the worst of both possible worlds!
unustable, giving it 2 more days before canceling the Pro
Trajectory ID: b53a6b86-2c3a-434e-
Error: HTTP 503 Service Unavailable
Sherlog:
TraceID: 0x1d182c27
Headers: {“Alt-Svc”:[“h3=“:443”; ma=2592000,h3-29=“:443”; ma=2592000”],“Content-Length”:[“421”],“Content-Type”:[“text/event-stream”],“Date”:[“Wed, 18 Mar 2026 12:39:14 GMT”],“Server”:[“ESF”],“Server-Timing”:[“gfet4t7; dur=215”],“Vary”:[“Origin”,“X-Origin”,“Referer”],“X-Cloudaicompanion-Trace-Id”:[“1d182c2763e2312e”],“X-Content-Type-Options”:[“nosniff”],“X-Frame-Options”:[“SAMEORIGIN”],“X-Xss-Protection”:[“0”]}
{
“error”: {
“code”: 503,
“details”: [
{
“@type”: “type.googleapis.com/google.rpc.ErrorInfo”,
“domain”: “cloudcode-pa.googleapis.com”,
“metadata”: {
“model”: “gemini-3-flash-agent”
},
“reason”: “MODEL_CAPACITY_EXHAUSTED”
}
],
“message”: “No capacity available for model gemini-3-flash-agent on the server”,
“status”: “UNAVAILABLE”
}
}
too many 503 now same at my side on flash and pro 3.1
Yeh, unusuable now Wed March 18
Yep, litterally getting this error now almost 9/10 requests. And had to press “retry” like 10 times untill it finally do something.
Nah man.
They complitely broke it.
Moving back to Claude again I guess.
its too irritating even on pro plan having same issue
Is broken for today for sure.
Impossible to work today. Every request take 4 times to click retry button.
Returning user of the Pro plan here..
It’s very sad to see the plan changing and limiting models for a week after just a couple of prompts. I’ve always enjoyed the plan mode here, but these new constraints make it hard to maintain any kind of momentum.
To make matters worse, the platform isn’t even working today. I just keep getting the same error message after a couple of seconds after submitting each prompt: Our servers are experiencing high traffic right now, please try again in a minute.
It’s becoming difficult to justify investing more time with Antigravity when the reliability just isn’t there anymore, especially since there are way more solid tools out there.
This is becoming ridiculous. I can’t even continue building my app or running tests because every few seconds, all I get is “Error: Our servers are experiencing high traffic right now, please try again in a minute.” over and over again, and I’m paying for this? I can’t renew my subscription again if this is what I’m getting.
Even on the Pro plan which I’m on, I can’t even enjoy the other models apart from Gemini Flash because, for some reason, once they are used up, it resets a week later, even though it’s supposed to be 5 hours. I specifically got the Pro plan because of this 5-hour reset duration, and I cant even get to experience it? So have I been lied to or what? Isn’t this false advertisement?
This is highly disappointing for Google.
We’re finding new ways to share more capacity with the Antigravity community. During windows where we have unused capacity, currently in the late afternoon PST, you’ll notice that your baseline quota stretches further. You’ll get more requests and less downtime within your existing plan.
As global demand shifts, we’ll continue to adjust these windows to pass any available capacity back to you. We’re committed to expanding access and ensuring every developer has the capacity they need to keep building.
Good afternoon. I am an Ultra subscriber, but the error “Our servers are experiencing high traffic right now, please try again in a minute” appears 15-20 times during a single model response… and this is on an Ultra subscription.
Trajectory ID: 3bed7275-bbbe-4297-a9d0-2b4d4f35bb27 Error: HTTP 503 Service Unavailable Sherlog: TraceID: 0x7b92ece46f666555 Headers: {"Alt-Svc":["h3=":443"; ma=2592000,h3-29=":443"; ma=2592000"],"Content-Length":["429"],"Content-Type":["text/event-stream"],"Date":["Wed, 18 Mar 2026 21:32:32 GMT"],"Server":["ESF"],"Server-Timing":["gfet4t7; dur=886"],"Vary":["Origin","X-Origin","Referer"],"X-Cloudaicompanion-Trace-Id":["7b92ece46f666555"],"X-Content-Type-Options":["nosniff"],"X-Frame-Options":["SAMEORIGIN"],"X-Xss-Protection":["0"]}
{ "error": { "code": 503, "details": [ { "@type": "type.googleapis.com/google.rpc.ErrorInfo", "domain": "cloudcode-pa.googleapis.com", "metadata": { "model": "claude-opus-4-6-thinking" }, "reason": "MODEL_CAPACITY_EXHAUSTED" } ], "message": "No capacity available for model claude-opus-4-6-thinking on the server", "status": "UNAVAILABLE" } }
Can anything be done about this?
It was the same on my Ultra Account which is now suspended for using OpenCode… You cannot do anything with the Subscription