17 Nov 21:53 UTCIdentifiedThe issue has been identified and a fix is being implemented. Current impact: predictions on OpenAI models are back to normal. Predictions on other models may still return HTTP 500 errors or stay in a processing state while we continue to work on a fix.
17 Nov 16:32 UTCIdentifiedThe issue has been identified and a fix is being implemented. Current impact: prediction requests for some models may return HTTP 500 errors, and some predictions may stay in a processing state. Customers using the API may also see validation errors on responses.
17 Nov 15:02 UTCInvestigatingWe are currently investigating this issue. Cloudflare is aware of, and investigating, an issue affecting predictions for some models on Replicate. Customers may see prediction requests return HTTP 500 errors, and some predictions may remain in a processing state without completing. Some API responses are also missing version information, which can cause validation errors in client libraries. We are working to understand the full impact and mitigate this problem.
Around it
From the sources ChangeIntel reads; a match in time is not a cause
Up to 100 recent field edits are shown; older edits may be outside retention.
17 Nov 2026 · 21:55:14 UTCSummaryThe issue has been identified and a fix is being implemented. Current impact: prediction requests for some models may return HTTP 500 errors, and some predictions may stay in a processing state. Customers using the API may also see validation errors on responses. → The issue has been identified and a fix is being implemented. Current impact: predictions on OpenAI models are back to normal. Predictions on other models may still return HTTP 500 errors or stay in a processing state while we continue to work on a fix.
17 Nov 2026 · 16:36:46 UTCSummaryWe are currently investigating this issue. Cloudflare is aware of, and investigating, an issue affecting predictions for some models on Replicate. Customers may see prediction requests return HTTP 500 errors, and some predictions may remain in a processing state without completing. Some API responses are also missing version information, which can cause validation errors in client libraries. We are working to understand the full impact and mitigate this problem. → The issue has been identified and a fix is being implemented. Current impact: prediction requests for some models may return HTTP 500 errors, and some predictions may stay in a processing state. Customers using the API may also see validation errors on responses.