Antigravity Claude Opus 4.6 Thinking — persistent 503 MODEL_CAPACITY_EXHAUSTED (ongoing several hours). Ultra Sub

Google Antigravity
1

I’ve been hitting repeated 503s on claude-opus-4-6-thinking in Antigravity for several hours now. the errors were happening more frequently over the last few days. tonight, every request fails with the same capacity-exhausted error, so the model is effectively unusable for me right now.
I tried switching off thinking, changing to sonnet, same issue

Model: claude-opus-4-6-thinking

  • Error: HTTP 503 UNAVAILABLE — MODEL_CAPACITY_EXHAUSTED

  • Domain: cloudcode-pa.googleapis.com

  • Trajectory ID: b1258b0f-edab-47ba-8619-b00f8603

  • Trace ID: 0x9cd99dceed67

  • Timestamp: 2026-04-08 ~08:58 GMT (issue ongoing all night prior)

  • Frequency: every request

    {
    “error”: {
    “code”: 503,
    “message”: “No capacity available for model claude-opus-4-6-thinking on the server”,
    “status”: “UNAVAILABLE”,
    “details”: [{
    “reason”: “MODEL_CAPACITY_EXHAUSTED”,
    “domain”: “cloudcode-pa.googleapis.com”,
    “metadata”: { “model”: “claude-opus-4-6-thinking” }
    }]
    }
    }

3 Likes
2

antigrav is not getting 503’s with Gemini. it’s just getting the errors with Claude models.

3

hmm, are you using OC with AntiGravity sub? Because turning off thinking is not really possible on AG. Also how did you get the detailed error message.

4

Trajectory ID: 5e282278-d5ca-400e-9e6
Error: HTTP 503 Service Unavailable
Sherlog:
TraceID: 0x2cb1de82d
Headers: {“Alt-Svc”:[“h3=“:443”; ma=2592000,h3-29=“:443”; ma=2592000”],“Content-Length”:[“429”],“Content-Type”:[“text/event-stream”],“Date”:[“Wed, 08 Apr 2026 11:52:56 GMT”],“Server”:[“ESF”],“Server-Timing”:[“gfet4t7; dur=1592”],“Vary”:[“Origin”,“X-Origin”,“Referer”],“X-Cloudaicompanion-Trace-Id”:[“2cb1de82d632cc92”],“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”
}
}

5

I’m having the same issue. The number of errors has been increasing every day for about three days now. I have an AI Ultra for Business account and using heavily Gemini, but with that errors is unusable.

I was switching with my personal AI Pro account, but that account has having the same issue with the model capacity error.

6

I meant planning or fast for conversation mode.

it’s an option when you get the agent error.

7

i went past my usage limit even on ultra and dipped into personal credits… did that happen to you at all? wondering if thats a trigger

8

posting an update

  • 22 hours later

Getting errors 1 every 2-3 minutes. It’s not fully failing though.

Trajectory ID: c90b9126-e403-4aff-aefc-3f347c146
Error: HTTP 503 Service Unavailable
Sherlog:
TraceID: 0xb15e3cd1d9496
Headers: {“Alt-Svc”:[“h3=“:443”; ma=2592000,h3-29=“:443”; ma=2592000”],“Content-Length”:[“429”],“Content-Type”:[“text/event-stream”],“Date”:[“Thu, 09 Apr 2026 06:37:21 GMT”],“Server”:[“ESF”],“Server-Timing”:[“gfet4t7; dur=293”],“Vary”:[“Origin”,“X-Origin”,“Referer”],“X-Cloudaicompanion-Trace-Id”:[“b15e3cd1d9496bbf”],“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”
}
}

9

hour later

have started getting multiple errors per request… and not making it through a single request

10

Ultra too, its insane, not useable for days now…
I pay for nothing I feel

11

it’s 2am est…
day 3…
cannot use AG at all. error instantly on retry

12

Trajectory ID: 18b1f57f-59cb-421c-a4e0-e5e5d4a81
Error: HTTP 400 Bad Request
Sherlog:
TraceID: 0x9fb07fa31c642
Headers: {“Alt-Svc”:[“h3=“:443”; ma=2592000,h3-29=“:443”; ma=2592000”],“Content-Length”:[“382”],“Content-Type”:[“text/event-stream”],“Date”:[“Fri, 10 Apr 2026 06:18:38 GMT”],“Server”:[“ESF”],“Server-Timing”:[“gfet4t7; dur=163”],“Vary”:[“Origin”,“X-Origin”,“Referer”],“X-Cloudaicompanion-Trace-Id”:[“9fb07fa31c64285f”],“X-Content-Type-Options”:[“nosniff”],“X-Frame-Options”:[“SAMEORIGIN”],“X-Xss-Protection”:[“0”]}

{
“error”: {
“code”: 400,
“details”: [
{
“@type”: “type.googleapis.com/google.rpc.BadRequest”,
“fieldViolations”: [
{
“description”: “Invalid project resource name”,
“field”: “name”
}
]
}
],
“message”: “Invalid project resource name projects/”,
“status”: “INVALID_ARGUMENT”
}
}

different error though now?

13

“…and dipped into personal credits…”

same! never had this problem until i used credits