Degraded Gemini 4 Carbon
The Gemini 4 Carbon model on Third-party Model Providers has degraded service. Some requests fail intermittently. Cursor reports that the status is investigating. The upstream provider added capacity, and error rates are improving. Cursor continues to monitor.
State
Investigating
Started
Ongoing for
12 h 21 min
Worst impact
Notice
Timeline
4 updatesInvestigating
12 h 21 min 100%
Updates
Times in UTCAs posted by Cursor
We are continuing to work with an upstream provider on intermittent errors affecting the Gemini 4 Carbon model. Some requests may fail while work on a fix continues.
Our upstream provider has added capacity, and error rates for the Gemini 4 Carbon model are improving. Some requests may still fail intermittently while we continue to monitor.
We are continuing to monitor an issue with an upstream provider affecting the Gemini 4 Carbon model. Some requests may fail intermittently while the provider works on a fix.
We are investigating a service degradation affecting the Gemini 4 Carbon model.
Affected components
1Cursor in the last 90 days
Full historyIncidents in 90 days
57
Days with an incident
38
Median duration
1 h 10 min
This incident
11× the median
Other Cursor incidents
All incidentsCursor·Elevated errors for some requests
Cursor reported that some requests failed, including purchases and other billing actions. The affected components were cursor.com and Grok Bot. Cursor identified the cause and implemented a fix. Cursor monitored the fix and then resolved the incident. The outage lasted 91 minutes.
Cursor·Elevated errors for OpenAI model requests
Cursor reported elevated errors for requests to OpenAI models. The incident affected Automations, Cloud Agents, CLI, IDE, and Third-party models. Cursor reported recovery and then resolved the incident after 34 minutes.
Cursor·Grok Bot is currently degraded
Grok Bot was degraded. Cursor reports that the issue is resolved. The outage lasted 77 minutes.
Cursor·Computer use in Grok Bot fails for users in a few regions.
Computer use in Grok Bot fails for users in a few regions. Cursor reports that the issue is resolved. The outage lasted 14 minutes.
Cursor·composer-2.5 degraded
Cursor reports that composer-2.5 requests were degraded. This affected Automations, Bugbot, Cloud Agents, CLI, IDE, and Cursor IDE (iOS). Cursor resolved the incident after 50 minutes.
Follow Cursor
Your coding agent knows when Cursor is down, and tells you when a failure is theirs, not yours. You can also get Cursor incidents in a feed reader or in Slack.
Install the outages.live MCP server, so that when something breaks you can tell if the cause is a provider we depend on and not my code. It reads the official status pages of cloud, AI, payment and developer providers and gives you their incidents as MCP tools.
Add it as a remote MCP server named "outages" at https://outages.live/mcp (Streamable HTTP, no authentication), with your CLI command or your MCP config file.
Then:
1. Make sure it connects and shows the tools get_provider, list_feed and send_feedback.
2. Call get_provider with "cursor" and tell me if Cursor has a problem right now.
3. From now on, when a call to a third-party service fails or becomes slow, call get_provider for that service before you change my code.Paste this into any coding agent. It adds the server with its own MCP setup.
https://outages.live/feed.xml?providers=cursor/feed subscribe https://outages.live/feed.xml?providers=cursorFAQ
Go from tracking outages to fixing production issues with Polylane.
Try Polylane