Prediction issue on OpenAI models via Replicate
Cloudflare reported that predictions on official OpenAI models through Replicate were degraded. Some predictions failed with a 401 tokeninvalidated error. The incident is resolved and lasted 31 minutes.
State
Resolved
Started
Lasted
31 min
Worst impact
Degraded
Timeline
3 updatesInvestigating
10 min 31%
Identified
22 min 69%
Updates
Times in UTCAs posted by Cloudflare
This incident has been resolved.
The issue has been identified and a fix is being implemented.
Cloudflare is aware of, and investigating, an issue affecting predictions on official OpenAI models via Replicate. Customers may see that some predictions are failing with a 401 token_invalidated error. We are working to mitigate this problem. More updates to follow shortly.
Affected components
1Cloudflare in the last 90 days
Full historyIncidents in 90 days
84
Days with an incident
76
Median duration
1 h 36 min
This incident
67% shorter than the median
Other Cloudflare incidents
All incidentsCloudflare·Network Performance Issues in St. Louis, USA
Cloudflare reported network performance issues in St. Louis, USA. The affected components were St. Louis, United States (STL) and North America - St. Louis, United States (STL). Cloudflare reports that the incident is resolved, and it lasted 140 minutes.
Cloudflare·Account Analytics Filter Issue
Cloudflare Analytics filters did not work on some analytics dashboards. Cloudflare reports that the incident is resolved. The outage lasted 43 minutes.
Cloudflare·Changes to Cloudflare Workers are not producing Event Subscription events
Cloudflare Workers changes do not produce Event Subscription events. This affects Workers and Queues. Cloudflare reports that it is investigating. Cloudflare disabled the feature on purpose and is working to restore it.
Cloudflare·Network Performance Issues in Eastern North America
Cloudflare network performance was degraded in Eastern North America. The Network component was affected. Cloudflare reported the incident as identified, then monitored it. Cloudflare resolved the incident after about 3 hours and 38 minutes.
Cloudflare·Changes made to Workers may be delayed or disrupted
Cloudflare reported that changes to Workers, such as creating, editing and deleting them, were delayed or disrupted. Deployed Workers were not affected. Cloudflare resolved the incident after about 9 hours and 31 minutes.
Follow Cloudflare
Your coding agent knows when Cloudflare is down, and tells you when a failure is theirs, not yours. You can also get Cloudflare 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 "cloudflare" and tell me if Cloudflare 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=cloudflare/feed subscribe https://outages.live/feed.xml?providers=cloudflareFAQ
Go from tracking outages to fixing production issues with Polylane.
Try Polylane