No capacity available for model gemini-3.1-pro-high on the server"

AG’s stability has been very poor these past few days. After logging in and out multiple times, I can no longer use the Gemini series. Is the official team implementing stricter regional traffic restrictions?

Trajectory ID: ZZZZZ
Error: HTTP 503 Service Unavailable
Sherlog:
TraceID: zzzzzz
Headers: {“Alt-Svc”:[“h3=“:443”; ma=2592000,h3-29=“:443”; ma=2592000”],“Content-Length”:[“419”],“Content-Type”:[“text/event-stream”],“Date”:[“Tue, 14 Apr 2026 21:16:26 GMT”],“Server”:[“ESF”],“Server-Timing”:[“gfet4t7; dur=415”],“Vary”:[“Origin”,“X-Origin”,“Referer”],“X-Cloudaicompanion-Trace-Id”:[“bcc221c1302318e9”],“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-3.1-pro-high”
},
“reason”: “MODEL_CAPACITY_EXHAUSTED”
}
],
“message”: “No capacity available for model gemini-3.1-pro-high on the server”,
“status”: “UNAVAILABLE”
}

Im having that issue right now

It appears to be a widespread outage as there are several other related threads appearing. The service is down overall

It’s just a bummer! I resumed playing around with MCPs on AG and then this Error (it’s also on gemini-3-flash-agent as well)

A few weeks back I tried tried to test the Google Drive MCP and ran into some other JSON parse error.

There needs to be better error reporting, because otherwise its confusing whether the problem is on the User’s end or Google Server end.

I’ve not had a single productive experience with Antigravity.

Encontre la solucion al error de limite, instale de nuevo la version del 22 de febrero y funciona con todos los modelos, mientras usare esa version mientras haya una solución. Si alguien la necesita, escibanme.

I’m also having the same issue.

Parece que el bug es con un proposito, llevaba 40 min funcionando bien con la version anterior, y de repente como que se dio cuenta que era cuenta ultra y mando de nuevo el error

It’s all down. Nothing is working. I would blame the storms, but I’m not gonna assume that.

Hopefully this means the quota issues will be fixed soon but… let’s not get our hopes up :laughing:

What’s going on with Gemini 3.1 Pro over the past couple of days? I haven’t been able to use it at all. Every time I try, it just gets stuck on “loading/working” and never responds. I’ve literally waited 10+ minutes staring at the screen with nothing happening.

What’s even more frustrating is that it still consumes my usage/quota without giving any response.

This is seriously affecting productivity. Is this a known issue?

I’ve also encountered this issue. It’s been going on for two hours, and all models—including Gemini 3.1 Flash—are showing this error:
“Our servers are experiencing high traffic right now, please try again in a minute.”

still not working for all model since few hours ago. please fix it ASAP….

Me too, Hope fixing it as soon as possible

The same error. I quick. I will change to claude code.

The issue has lasted for over six hours.

I’m switching to opencode configured with Grok and a local LM Studio running on a mac mini. I can’t continue to wait for google to unbork itself.

My Ultra membership rarerly works, its always this capacity issues. Barely able to use it ever

Our servers are experiencing high traffic right now, please try again in a minute.

So google with a big data center still have this problem? lmao

I also have same issue! Even the model quotation of Ultra plan is full (not yet used today)!

i have the same issue on free plan. yet to try the program yet as it hasn’t worked once in two weeks.