Antigravity trigger Model quota limit weekly cooldown after version update

I am experiencing the same issue. Antigravity says quota limit reached and even though I have AI Pro I see refresh windows for free users. I got one response out of Claude 4.5 Opus (yes I am sure that I had Claude and not Gemini selected) and then the quota limit blocked me.

What happened (actual)

2026-01-24 21:46:36.769 [info] [Window] Started local extension host with pid 12873.
2026-01-24 21:46:37.365 [error] [Window] [Extension Host] Failed to update user status Error: LanguageServerClient must be initialized first!
2026-01-24 21:46:37.770 [info] (Antigravity) 2026-01-24 21:46:37.770 [INFO]: Language server started
2026-01-24 21:46:37.774 [info] W0124 21:46:37.764515 12909 client.go:137] failed to set auth token
2026-01-24 21:46:37.774 [info] W0124 21:46:37.765235 12909 log_context.go:106] Cache(userInfo): Singleflight refresh failed: You are not logged into Antigravity.
2026-01-24 21:46:37.774 [info] I0124 21:46:37.773667 12909 server.go:1504] initialized server successfully in 222.073667ms

  • (Extension runtime starts.)

  • (Init-order bug: extension tries to use LS client before it’s ready → breaks auth/user-status wiring.)

  • (LS comes up after the extension already failed.)

  • (Auth token never gets set in LS → treated as not logged in.)

  • (User/plan fetch fails due to missing token; “singleflight” just dedupes concurrent refreshes.)

  • (LS is running, but unauthenticated.)


How it should be (expected)

2026-01-24 21:46:36.769 [info] [Window] Started local extension host with pid 12873.
2026-01-24 21:46:37.000 [info] [Window] [Extension Host] Initializing LanguageServerClient...
2026-01-24 21:46:37.200 [info] (Antigravity) 2026-01-24 21:46:37.200 [INFO]: Language server started
2026-01-24 21:46:37.250 [info] [Window] [Extension Host] LanguageServerClient initialized
2026-01-24 21:46:37.300 [info] [Window] [Extension Host] Authenticated session detected; setting auth token
2026-01-24 21:46:37.320 [info] I0124 21:46:37.320000 12909 client.go:137] auth token set
2026-01-24 21:46:37.350 [info] I0124 21:46:37.350000 12909 log_context.go:106] Cache(userInfo): refresh succeeded
2026-01-24 21:46:37.380 [info] [Window] [Extension Host] Starting update loop (user status / available models)

  • (Extension runtime starts.)

  • (LS client init begins; nothing that needs LS client should run yet.)

  • (LS process starts.)

  • (LS client is connected/ready.)

  • (Extension now propagates token to LS.)

  • (LS acknowledges token; opposite of “failed to set auth token”.)

  • (User/plan info becomes available.)

  • (Only now start polling loops like user status / models.)

See: Fix for Quota Limits - OAuth issue

1 Like

Thank you for bringing this to our attention. We sincerely apologize for the inconvenience this has caused. We have escalated the issue to our internal teams for a thorough investigation. Any additional context or screenshots you can provide would be invaluable as we work toward a resolution.

5 Likes

Hello, I was really impressed with Antigravity and the use of the Claude Opus model until now. I reached the limit and waited three days because the limit was locked until January 25, 2026, at 6:20:57 PM. Then I logged in and saw that the limit had been extended until February 1, 2026, at 6:20:57 PM. I’m a Google AI Pro user, and this can’t be right. Is Antigravity broken?

2 Likes

I have the exact error logs that @samiralibabic has pointed out. You can reffer to his post here: Fix for Quota Limits - OAuth issue

if specific details you would to know let me know.

Just a question, if you have any idea. It seems The error has been happening for a while now and still does, How come the quota limits wasn’t always flagged and why is the flag occurs on some models and not all?

Same issue here as well.

In my case, the problem has been persisting for several days now. The quota reset date was initially set to January 27, but instead of being restored, it has now been pushed back by another 7 days starting today.

Bump. Bump.
This is a serious issue!! I’m locked out!! Kindly escalate this issue for fixing promptly @Google Team.

2 Likes

Please don’t spread misinformation; I reported a bug but they said it was intentional and needed to greatly reduce the weekly limit due to the “model terminated due to error” issue which is caused by overloaded capacity. Hopefully, this extreme rate limit is temporary

The people on internal teams have said it’s necessary and intentional, not a bug

Where did you get that from?

Same issue here, this has been going on for days with no updates or bug fixes

exact same date happened here

It worked perfectly well, but in January what started happening, was that at first there were many failures on Anthropic models, so even when at failure, it should not count toward the quota, what happened for me, was 10 errors in succession for Anthropic models, then a sudden quota hit of 4-7 days, after the models literally made no outputs. Then I bought another subscription, it hit the quota after 4 hours, even when before i could help myself with coding for 8-10 hours with Sonnet, without hitting the quota. So not only the quality of the outputs went down, but now instead of daily quota it is maybe 4-10 day quota. Absolutely ruined costumer experience. So now im paying for 2 x Pro accounts not being able to use the service.

Do you have some proof, or is this a ‘trust me bro’ kind of scenario?

It would be good if everyone could post here about the issue:

https://support.google.com/googleone/?hl=en#topic=9171059

Google also prioritizes issues based on the number of reports.

2 Likes

I’m having the same issue and only realized it after coming here.

2026-01-27 06:00:54.819 [info] W0127 06:00:54.819124 28936 log_context.go:106] Cache(userInfo): Singleflight refresh failed: You are not logged into Antigravity.

2026-01-27 06:00:54.820 [info] W0127 06:00:54.820156 28936 client.go:137] failed to set auth token

2026-01-27 06:00:54.820 [info] W0127 06:00:54.820156 28936 log_context.go:106] Cache(userInfo): Singleflight refresh failed: You are not logged into Antigravity.

2026-01-27 06:00:54.822 [info] W0127 06:00:54.822225 28936 client.go:137] failed to set auth token

2026-01-27 06:00:54.822 [info] W0127 06:00:54.822225 28936 log_context.go:106] Cache(userInfo): Singleflight refresh failed: You are not logged into Antigravity.

Waited a week to us Claude, open this AM and it immediately pushed out.

1 Like

What I have been doing is using the precious Claude token limit to write my detailed feature specs. Taking into account the framework and rules of my environment. I also have Claude include a full section in the feature doc for a future coding agent to use, the rules of the environment and how to make the changes. Then feed that to gemini flash. It cuts down on hallucinations and gives the gemini models more info that a typical prompt. It is not 100%. Use Claude as your senior architect making specs for the devs…gemini. You will have some cleanup but nothing like everyone has been experiencing.

Wes

2 Likes

It’s better if we wait for google entire team announcement for the quotas update instead of getting ahead of events