Still broken for me. As an Ultra user, it is incredibly heartbreaking for it to be down this long.
Still down 
I have ultra as well… all models down
Use alternate account if you have one) free tier is working pretty well
P.S. I have a PRO and no Antigravity for me too
I had the same problem but now it seems it started work again.
I uninstalled and cleaned all the residual configuration files (I used IoBit Uninstaller to run a power uninstall, but it’s full of bloatware so be careful, surely there’s some better alternative), then installed agai as a fresh install (without importing my vs code settings)
I’ve had the same issue on Mac starting after the update yesterday. I did a clean uninstall and deleted all hidden files in my root and that fixed it for a few hours but now it is back and I am at a standstill again.
This is not a local issue this is a known issue affecting both Pro/Ultra plan users as well as cloud api users it seems.
This is what gemini itself via web has to say:
It is a complete farce, and the reason you aren’t finding a polished “official announcement” is part of the problem. Instead of a single press release, Google has acknowledged the crisis in fragments across developer channels.
The most “official” explanation for your hasAnthropicModelAccess: false flag and the total lockout is a forced backend migration occurring right now (January 2026) as Google attempts to decouple AI Studio billing from the Google Cloud Console.
What Google Has Actually Admitted (Jan 2026)
-
The “Migration” Bug: Google product leads have confirmed that AI Studio is moving to a “Direct Billing” model to simplify payments. This transition has accidentally “orphaned” thousands of Google One Ultra subscribers, leaving them stuck in a loop where the system sees their payment but refuses to provision the “Paid” flags (like Claude access) to their developer tools.
-
The “Phantom Quota” Acknowledgment: In the official Google AI Developers Forum, employeesare actively replying to users with your exact symptoms, admitting the system is “treating paid subscribers as Free Tier users”.
-
Antigravity “Agent Termination”: As of January 15, 2026 (today), there is a massive spike in reports of “Agent execution terminated due to error”. Users on Ultra plans are reporting that their accounts are a “total blocker,” while—insultingly—free accounts on the same machines work perfectly.
-
The 7-Day Lockout: Google has acknowledged a bug where the intended 5-hour quota refresh for Pro/Ultra users has been erroneously extended to 4–7 days. This explains why you “ran out of credits” and are seeing resume dates set for next week (e.g., Jan 16-20).
Hello everyone!
Thank you for letting us know. We have identified this as a bug and escalated it internally. If you are able to provide further context or images, it would be incredibly helpful for our team as we work to resolve this.
We highly recommend filing bug reports directly through the Antigravity in-app feedback tool. You can do this by navigating to the top-right corner of the interface, clicking the Feedback icon, and selecting Report Issue.
On my side it worked this morning but just stopped working again now after only a few hours… 
2026-01-15 15:54:35.652 [info] 2026/01/15 15:54:35 maxprocs: Leaving GOMAXPROCS=8: CPU quota undefined
2026-01-15 15:54:35.694 [info] I0115 15:54:35.694234 18908 server.go:1172] Starting language server process with pid 18908
2026-01-15 15:54:35.695 [info] I0115 15:54:35.694731 18908 server.go:286] Setting GOMAXPROCS to 4
2026-01-15 15:54:35.698 [info] I0115 15:54:35.698207 18908 server.go:493] Language server will attempt to listen on host 127.0.0.1
2026-01-15 15:54:35.699 [info] I0115 15:54:35.698704 18908 server.go:508] Language server listening on random port at 55487 for HTTPS
2026-01-15 15:54:35.699 [info] I0115 15:54:35.698704 18908 server.go:515] Language server listening on random port at 55488 for HTTP
2026-01-15 15:54:35.792 [info] I0115 15:54:35.792574 18908 server.go:486] Created extension server client at port 55486
2026-01-15 15:54:35.859 [info] W0115 15:54:35.859127 18908 client.go:136] failed to set auth token
2026-01-15 15:54:35.867 [info] (Antigravity) 2026-01-15 15:54:35.867 [INFO]: Language server started
2026-01-15 15:54:35.872 [info] W0115 15:54:35.859625 18908 log_context.go:106] Cache(userInfo): Singleflight refresh failed: You are not logged into Antigravity.
2026-01-15 15:54:35.873 [info] I0115 15:54:35.872614 18908 server.go:1528] initialized server successfully
2026-01-15 15:54:36.014 [info] 2026/01/15 15:54:36 http: TLS handshake error from 127.0.0.1:55504: read tcp 127.0.0.1:55487->127.0.0.1:55504: wsarecv: An established connection was aborted by the software in your host machine.
2026-01-15 15:54:36.042 [info] (Antigravity) 2026-01-15 15:54:36.042 [INFO]: LS lspClient started successfully
2026-01-15 15:54:43.391 [info] E0115 15:54:43.390756 18908 log.go:354] encountered conflicting override on experiment: cascade-include-ephemeral-message
2026-01-15 15:54:43.391 [info] E0115 15:54:43.391752 18908 log.go:354] encountered conflicting override on experiment: cascade-knowledge-config
2026-01-15 15:54:43.503 [info] I0115 15:54:43.502506 18908 planner_generator.go:273] Requesting planner with 6 chat messages at model retry attempt 1 and API retry attempt 1
2026-01-15 15:54:48.648 [info] E0115 15:54:48.647790 18908 log.go:362] 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:[139] Content-Type:[text/event-stream] Date:[Thu, 15 Jan 2026 21:54:44 GMT] Server:[ESF] Server-Timing:[gfet4t7; dur=120] Vary:[Origin X-Origin Referer] X-Cloudaicompanion-Trace-Id:[306ae459e1464972] X-Content-Type-Options:[nosniff] X-Frame-Options:[SAMEORIGIN] X-Xss-Protection:[0]] 0x123811f0b680 139 [] false false map[] 0x123811ec1540 0x1237ce2f6b40}: 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:[139] Content-Type:[text/event-stream] Date:[Thu, 15 Jan 2026 21:54:44 GMT] Server:[ESF] Server-Timing:[gfet4t7; dur=120] Vary:[Origin X-Origin Referer] X-Cloudaicompanion-Trace-Id:[306ae459e1464972] X-Content-Type-Options:[nosniff] X-Frame-Options:[SAMEORIGIN] X-Xss-Protection:[0]] 0x123811f0b680 139 [] false false map[] 0x123811ec1540 0x1237ce2f6b40}
2026-01-15 15:54:48.649 [info] I0115 15:54:48.649284 18908 planner_generator.go:273] Requesting planner with 7 chat messages at model retry attempt 1 and API retry attempt 2
2026-01-15 15:54:58.788 [info] E0115 15:54:58.788483 18908 log.go:362] 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:[139] Content-Type:[text/event-stream] Date:[Thu, 15 Jan 2026 21:54:49 GMT] Server:[ESF] Server-Timing:[gfet4t7; dur=113] Vary:[Origin X-Origin Referer] X-Cloudaicompanion-Trace-Id:[56fb46f7cf28c50a] X-Content-Type-Options:[nosniff] X-Frame-Options:[SAMEORIGIN] X-Xss-Protection:[0]] 0x1237ee1b1280 139 [] false false map[] 0x1237fd3f7180 0x1237ce2f6b40}: 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:[139] Content-Type:[text/event-stream] Date:[Thu, 15 Jan 2026 21:54:49 GMT] Server:[ESF] Server-Timing:[gfet4t7; dur=113] Vary:[Origin X-Origin Referer] X-Cloudaicompanion-Trace-Id:[56fb46f7cf28c50a] X-Content-Type-Options:[nosniff] X-Frame-Options:[SAMEORIGIN] X-Xss-Protection:[0]] 0x1237ee1b1280 139 [] false false map[] 0x1237fd3f7180 0x1237ce2f6b40}
2026-01-15 15:54:58.790 [info] I0115 15:54:58.789941 18908 planner_generator.go:273] Requesting planner with 8 chat messages at model retry attempt 1 and API retry attempt 3
2026-01-15 15:54:58.924 [info] E0115 15:54:58.923615 18908 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:[139] Content-Type:[text/event-stream] Date:[Thu, 15 Jan 2026 21:54:59 GMT] Server:[ESF] Server-Timing:[gfet4t7; dur=98] Vary:[Origin X-Origin Referer] X-Cloudaicompanion-Trace-Id:[e8017411708617eb] X-Content-Type-Options:[nosniff] X-Frame-Options:[SAMEORIGIN] X-Xss-Protection:[0]] 0x1237db3c4180 139 [] false false map[] 0x1237db325540 0x1237ce2f6b40}: 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:[139] Content-Type:[text/event-stream] Date:[Thu, 15 Jan 2026 21:54:59 GMT] Server:[ESF] Server-Timing:[gfet4t7; dur=98] Vary:[Origin X-Origin Referer] X-Cloudaicompanion-Trace-Id:[e8017411708617eb] X-Content-Type-Options:[nosniff] X-Frame-Options:[SAMEORIGIN] X-Xss-Protection:[0]] 0x1237db3c4180 139 [] false false map[] 0x1237db325540 0x1237ce2f6b40}: model api cannot be reached
Same issue here for the last couple of hours using the Ultra plan. I have sent feedback via the in-app interface but would love to understand what is causing the issue and time for resolution
Trigger failed: [[claude-opus-4-5-thinking]]: ERROR API request failed (streamGenerateContent): Cloud Code stream failed (429) (4476ms) [[claude-sonnet-4-5-thinking]]: ERROR API request failed (streamGenerateContent): Cloud Code stream failed (429) (4386ms) [[gemini-3-flash]]: ERROR API request failed (streamGenerateContent): Cloud Code stream failed (429) (4435ms) [[gemini-3-pro-high]]: ERROR API request failed (streamGenerateContent): Cloud Code stream failed (429) (4372ms) [[gemini-3-pro-low]]: ERROR API request failed (streamGenerateContent): Cloud Code stream failed (429) (4234ms) [[gemini-3-pro-image]]: ERROR API request failed (streamGenerateContent): Cloud Code stream failed (429) (4208ms) [[gpt-oss-120b-medium]]: ERROR API request failed (streamGenerateContent): Cloud Code stream failed (429) (4346ms)
omg i’m glad i found this thread. This has been driving me nuts the last 12-24 hours. now its just dead in the water. Tried so many things. 429 ratelimited. TLS errors. auth errors. Deleting all system cached data and reinstalling didn’t fix it either.
On Ultra plan. Cannot work at all.
@Abhijit_Pramanik Doesn’t look like i can even submit bug reports.
But heres my server log:
agent terminated due to error.
Auth, tls and 429 errors
2026-01-16 11:29:58.627 [info] 2026/01/16 11:29:58 maxprocs: Leaving GOMAXPROCS=24: CPU quota undefined
2026-01-16 11:29:58.648 [info] I0116 11:29:58.648243 27827 server.go:1172] Starting language server process with pid 27827
2026-01-16 11:29:58.648 [info] I0116 11:29:58.648268 27827 server.go:286] Setting GOMAXPROCS to 4
2026-01-16 11:29:58.649 [info] I0116 11:29:58.649553 27827 server.go:493] Language server will attempt to listen on host 127.0.0.1
2026-01-16 11:29:58.649 [info] I0116 11:29:58.649696 27827 server.go:508] Language server listening on random port at 39783 for HTTPS
2026-01-16 11:29:58.650 [info] I0116 11:29:58.649717 27827 server.go:515] Language server listening on random port at 39109 for HTTP
2026-01-16 11:29:58.980 [info] I0116 11:29:58.980679 27827 server.go:486] Created extension server client at port 46607
2026-01-16 11:29:58.987 [info] W0116 11:29:58.986670 27827 client.go:136] failed to set auth token
2026-01-16 11:29:58.993 [info] (Antigravity) 2026-01-16 11:29:58.993 [INFO]: Language server started
2026-01-16 11:29:58.996 [info] W0116 11:29:58.987107 27827 log_context.go:106] Cache(userInfo): Singleflight refresh failed: You are not logged into Antigravity.
2026-01-16 11:29:58.997 [info] I0116 11:29:58.996547 27827 server.go:1528] initialized server successfully
2026-01-16 11:29:59.026 [info] 2026/01/16 11:29:59 http: TLS handshake error from 127.0.0.1:51168: remote error: tls: unknown certificate
2026-01-16 11:29:59.051 [info] (Antigravity) 2026-01-16 11:29:59.051 [INFO]: LS lspClient started successfully
2026-01-16 11:30:06.901 [info] I0116 11:30:06.901740 27827 planner_generator.go:273] Requesting planner with 5 chat messages at model retry attempt 1 and API retry attempt 1
2026-01-16 11:30:12.702 [info] E0116 11:30:12.702353 27827 log.go:362] 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:[139] Content-Type:[text/event-stream] Date:[Thu, 15 Jan 2026 22:30:07 GMT] Server:[ESF] Server-Timing:[gfet4t7; dur=766] Vary:[Origin X-Origin Referer] X-Cloudaicompanion-Trace-Id:[e959e483a40532cd] X-Content-Type-Options:[nosniff] X-Frame-Options:[SAMEORIGIN] X-Xss-Protection:[0]] 0x13e53d0fc8c0 139 false false map 0x13e53c513180 0x13e53d3ae9c0}: 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:[139] Content-Type:[text/event-stream] Date:[Thu, 15 Jan 2026 22:30:07 GMT] Server:[ESF] Server-Timing:[gfet4t7; dur=766] Vary:[Origin X-Origin Referer] X-Cloudaicompanion-Trace-Id:[e959e483a40532cd] X-Content-Type-Options:[nosniff] X-Frame-Options:[SAMEORIGIN] X-Xss-Protection:[0]] 0x13e53d0fc8c0 139 false false map 0x13e53c513180 0x13e53d3ae9c0}
2026-01-16 11:30:12.707 [info] I0116 11:30:12.707126 27827 planner_generator.go:273] Requesting planner with 6 chat messages at model retry attempt 1 and API retry attempt 2
2026-01-16 11:30:23.383 [info] E0116 11:30:23.383483 27827 log.go:362] 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:[139] Content-Type:[text/event-stream] Date:[Thu, 15 Jan 2026 22:30:13 GMT] Server:[ESF] Server-Timing:[gfet4t7; dur=635] Vary:[Origin X-Origin Referer] X-Cloudaicompanion-Trace-Id:[104428ae8c6e4253] X-Content-Type-Options:[nosniff] X-Frame-Options:[SAMEORIGIN] X-Xss-Protection:[0]] 0x13e53c683280 139 false false map 0x13e53bcb6140 0x13e53d3ae9c0}: 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:[139] Content-Type:[text/event-stream] Date:[Thu, 15 Jan 2026 22:30:13 GMT] Server:[ESF] Server-Timing:[gfet4t7; dur=635] Vary:[Origin X-Origin Referer] X-Cloudaicompanion-Trace-Id:[104428ae8c6e4253] X-Content-Type-Options:[nosniff] X-Frame-Options:[SAMEORIGIN] X-Xss-Protection:[0]] 0x13e53c683280 139 false false map 0x13e53bcb6140 0x13e53d3ae9c0}
2026-01-16 11:30:23.387 [info] I0116 11:30:23.387416 27827 planner_generator.go:273] Requesting planner with 7 chat messages at model retry attempt 1 and API retry attempt 3
2026-01-16 11:30:24.002 [info] E0116 11:30:24.001853 27827 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:[139] Content-Type:[text/event-stream] Date:[Thu, 15 Jan 2026 22:30:23 GMT] Server:[ESF] Server-Timing:[gfet4t7; dur=581] Vary:[Origin X-Origin Referer] X-Cloudaicompanion-Trace-Id:[9fdc370d5f785b95] X-Content-Type-Options:[nosniff] X-Frame-Options:[SAMEORIGIN] X-Xss-Protection:[0]] 0x13e53caf5240 139 false false map 0x13e53cc3fa40 0x13e53d3ae9c0}: 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:[139] Content-Type:[text/event-stream] Date:[Thu, 15 Jan 2026 22:30:23 GMT] Server:[ESF] Server-Timing:[gfet4t7; dur=581] Vary:[Origin X-Origin Referer] X-Cloudaicompanion-Trace-Id:[9fdc370d5f785b95] X-Content-Type-Options:[nosniff] X-Frame-Options:[SAMEORIGIN] X-Xss-Protection:[0]] 0x13e53caf5240 139 false false map 0x13e53cc3fa40 0x13e53d3ae9c0}: model api cannot be reached
Same blockage :”Agent terminated due to error” ! (Macbook; Google AI ULTRA).
Also just want to add as an Ultra plan user that has been suffering from this in the last two days. Started sometime in the afternoon US Central time and then lasted all the rest of the evening. woke up this morning, it went away, was able to work for about four or five hours and then once again was hit the same error across all models.
Pretty crazy for a $9 a DAY service. Verizon gave customers $20 credit for the month for 1/4th of a day down time. And I’m sure that average bill is less than $250 a month lol and across two days.
Plus zero public acknowledgement or support with zero idea of when this will be over. Antigravity has been incredibly useful, however I will keep this even in mind when considering switching back to Codex or Claude
Ultra plan SAME ISSUE
DOWN for 5 hours
Google FIX it ASAP!
Same here! I temporarily jumped back to Windsurf until Antigrvity is fixed.
Same, late afternoon Jan 14 I only got errors - it started working again morning of Jan 15 but now again repeated errors. Active Ultra subscription, stable internet, central time zone.
Got same issue on UltraPlan. No response from support by email after 24H. Gemini chatbot seems to think the issue is from a patch to prevent usage abuse that trigger on normal users. Resetting and reinstalling is a waste of time.
If you try on a different computer you’ll get same error, I tried. Only waiting for daily reset a 3am seams to work for some time. I got reblocked today after working 4h to reduce token usage in my project. .aignore file seems to help to minimise context windows. It`s like a gitignore but for AI context , you tell the agent to not read files types listed in there. I try my best to help, but google will need to fix this.
For me this is a nightmare. Losing X10 productivity for half a day twice in a row, that`s terrible. Currently notting come even close to antigravity for coding. Cursor terminal blindness in painfull to work with. I even tried running llm locally with vscode, the agent is completely blind on the local repo. DeepseekV3 is oversaturated. Google you’re our only hope …