High frequency of getting this error, unable to use
2 Likes
Same here, please fix it 
1 Like
I believe this is a load management process for google. I start to get the error and a couple of hours later it goes away. Interestingly, the first time this happened to me a couple of days ago, I just switched account and used the free plan. When that ran out, I want back to my paid plan and that worked.
Yesterday hitting the same agent problem I logged out to use another account on the free plan but then hit the authorisation problem the was raised by many people yesterday. I think google can identify that we have switched between plans so stopped this also to balance the load.
Will be interesting if it happens again today.
Gemini 3 Pro works well but Claude Opus 4.5 does not
I use Gemini 3 Pro works well but Claude Opus 4.5 not work
getting the error repeated since yesterday,.
1 Like
People are saying it works fine in pro plan yet I am in pro plan and same issue happening
5 Likes
Same here 
2 Likes
yes ultra is having same issue here
1 Like
For me here is the culprit !!! Good job !!
its almost unusable for at least two days. Constantly getting errors. This is absolutely unacceptable for users at Pro or Ultra plan.
Will probably ditch it and move to Claude Code plan until google gets their act together here.
1 Like
I am still getting this error (see attached image), I have an Ultra plan, and I tried using different model versions, after restarting my comp, i even completely removed and reinstalled Antigravity, and even with full automation, I keep on having to check if it stalled and press “retry” – this error seems to be worse with Opus, but it really is random (and expected) across the board with any model and task complexity. Help! This is frustrating not being to trust the loop! – can we just program the agent to automatically press the retry button 
3 Likes
The same here, PRO user.
1 Like
+1
It’s hard to work on the second day, I just click “repeat.” Ultra plan
2026-01-23 15:19:16.931 [info] E0123 15:19:16.930977 51016 log.go:362] agent executor error: request failed: &{429 Too Many Requests 429 HTTP/1.1 1 1 map[Alt-Svc:[h3=":443"; ma=2592000,h3-29=":443"; ma=2592000] Content-Length:[522] Content-Type:[text/event-stream] Date:[Fri, 23 Jan 2026 14:19:17 GMT] Server:[ESF] Server-Timing:[gfet4t7; dur=1909] Vary:[Origin X-Origin Referer] X-Cloudaicompanion-Trace-Id:[46cdd41d453cc3b4] X-Content-Type-Options:[nosniff] X-Frame-Options:[SAMEORIGIN] X-Xss-Protection:[0]] 0x1b87a73a6140 522 [] false false map[] 0x1b87e57b9540 0x1b875e0cfd40}: request failed: &{429 Too Many Requests 429 HTTP/1.1 1 1 map[Alt-Svc:[h3=":443"; ma=2592000,h3-29=":443"; ma=2592000] Content-Length:[522] Content-Type:[text/event-stream] Date:[Fri, 23 Jan 2026 14:19:17 GMT] Server:[ESF] Server-Timing:[gfet4t7; dur=1909] Vary:[Origin X-Origin Referer] X-Cloudaicompanion-Trace-Id:[46cdd41d453cc3b4] X-Content-Type-Options:[nosniff] X-Frame-Options:[SAMEORIGIN] X-Xss-Protection:[0]] 0x1b87a73a6140 522 [] false false map[] 0x1b87e57b9540 0x1b875e0cfd40}
On close to every possible interaction now, I am getting Agent terminated due to error. ALWAYS, it is caused by a 429 error from the backend. I am on the pro plan, and this bug occurs in close to every model I have tested.
same here. Unusable right now.
2026-01-23 17:33:08.604 [info] E0123 17:33:08.604247 12868 log.go:362] agent executor error: model unreachable: request failed: &{429 Too Many Requests 429 HTTP/1.1 1 1 map[Alt-Svc:[h3=“:443”; ma=2592000,h3-29=“:443”; ma=2592000] Content-Length:[436] Content-Type:[text/event-stream] Date:[Fri, 23 Jan 2026 14:33:11 GMT] Server:[ESF] Server-Timing:[gfet4t7; dur=245] Vary:[Origin X-Origin Referer] X-Cloudaicompanion-Trace-Id:[79e905c2c5bc31a0] X-Content-Type-Options:[nosniff] X-Frame-Options:[SAMEORIGIN] X-Xss-Protection:[0]] 0x3703738e4e40 436 [] false false map[] 0x3703768083c0 0x370370178d80}: request failed: &{429 Too Many Requests 429 HTTP/1.1 1 1 map[Alt-Svc:[h3=“:443”; ma=2592000,h3-29=“:443”; ma=2592000] Content-Length:[436] Content-Type:[text/event-stream] Date:[Fri, 23 Jan 2026 14:33:11 GMT] Server:[ESF] Server-Timing:[gfet4t7; dur=245] Vary:[Origin X-Origin Referer] X-Cloudaicompanion-Trace-Id:[79e905c2c5bc31a0] X-Content-Type-Options:[nosniff] X-Frame-Options:[SAMEORIGIN] X-Xss-Protection:[0]] 0x3703738e4e40 436 [] false false map[] 0x3703768083c0 0x370370178d80}: model api cannot be reached
Bought Ultra yesterday, worked fine. Today - basically unusable.
2 Likes
Same issue since 3-4 days, also the search_web tool is completely broken and unusable. I am being forced to rely on the browser sub-agent, which is extremely slow and inefficient.
This issue has been persisting for the last 3–4 days without any resolution, and there has been no response from the Antigravity team, which is frankly unacceptable for a production system.
On top of this, the Claude Opus models are repeatedly failing, making the platform unreliable for serious work.
This is a critical service degradation, and it needs immediate attention and acknowledgment.
7 Likes
Same error here, for a week… and premium plan, are here devs working on this? or will go to k*bygoogle?
1 Like