Live AI Providers Outage Feed
Track outages, degradations, and elevated error rates across the major AI providers in real time.
This week's ai providers outage roundup →- Degraded
Increase in users hitting Codex rate limits
Status: InvestigatingWe are investigating the issue for the listed services.Affected components Codex Web (Degraded performance) CLI (Degraded performance) App (Degraded performance) VS Code extension (Degraded performance) Codex API (Degraded performance)
- Degraded
Elevated errors on Claude Opus 4.7
May 22, 10:17 UTC Investigating - We are currently investigating this issue.
- Degraded
Elevated error rate
May 22, 04:16 UTCInvestigating - We are currently investigating this issue.
- Degraded
Elevated latency and error rates for ChatGPT 5.5 Thinking
Status: IdentifiedWe have identified that users are experiencing elevated latency and error rates for ChatGPT 5.5 Thinking. We are working on implementing a mitigation.Affected components Conversations (Degraded performance)
- Degraded
Prediction and Training status updates delayed
Status: IdentifiedOur message queues for prediction and training status updates are hitting capacity limits which are causing connection failures for queue consumers. We are in the process of bringing additional capacity online.Affected components Playground (Degraded performance) H100 Hardware (Degraded performance) HTTP API (Degraded performance) CPU Hardware (Degraded performance) Home Page (Degraded performance) L40S Hardware (Degraded performance) A100 Hardware (Degraded performance) Streaming API (Degraded performance) T4 Hardware (Degraded performance)
- Degraded
Elevated errors on Claude.ai
May 21, 20:24 UTCInvestigating - We are investigating elevated errors for users of Claude.ai, and will provide an update as soon as possible.
- Major
Constrained H100 capacity
Status: IdentifiedWe are seeing heightened demand for H100 hardware which is causing severe queue delays.Affected components H100 Hardware (Partial outage)
- Degraded
Elevated error rates on ChatGPT paid plans
Status: IdentifiedWe have identified that users are experiencing elevated errors for the impacted services. We are working on implementing a mitigation.Affected components Conversations (Degraded performance)
- Degraded
Elevated errors on Claude Opus 4.7
May 20, 14:43 UTCInvestigating - We are currently investigating this issue.
- Degraded
Elevated errors on Claude Haiku 4.5
May 20, 08:31 UTCMonitoring - A fix has been implemented and we are monitoring the results.May 20, 08:14 UTCInvestigating - We are currently investigating this issue.
- Degraded
API users may see increased error rates for GPT-5.4 and GPT-5.5
Status: MonitoringWe have applied the mitigation and are monitoring the recovery.Affected components Responses (Degraded performance) Chat Completions (Degraded performance)
- Degraded
Elevated errors on Claude Opus 4.7
May 19, 14:03 UTCInvestigating - We are currently investigating this issue.
- Degraded
Elevated errors on Claude Opus 4.7
May 19, 08:41 UTCInvestigating - We are currently investigating this issue.
- Degraded
Elevated errors on Claude Haiku 4.5
May 19, 07:00 UTCInvestigating - We are currently investigating this issue.
- Degraded
Elevated errors on Claude.ai and Claude Code login
May 19, 04:46 UTCMonitoring - From 04:34–04:42 UTC, we saw elevated errors and latency on requests within the Claude.ai applications. Additionally, users may have experienced errors when attempting to login, including authenticating Claude Code via Claude.ai. We are monitoring closely to ensure no further issues, and will provide an additional update shortly.
- Degraded
Elevated errors for Claude Opus 4.7
May 18, 22:19 UTCResolved - The issue affecting Claude services has been resolved. Impact occurred from 21:47 to 22:18 UTC.May 18, 22:10 UTCInvestigating - We are investigating elevated errors on requests to Claude Opus 4.7. We will provide an update as soon as possible.
- Degraded
Elevated errors on Claude Haiku 4.5
May 18, 06:12 UTCInvestigating - We are currently investigating this issue.
- Degraded
Elevated error rates on requests to multiple models
May 16, 18:18 UTCMonitoring - A fix has been implemented and we are monitoring the results.May 16, 18:08 UTCInvestigating - We are currently investigating this issue.
- Info
EU Residency: Scribe V1 downtime
Status: ResolvedThe issue has been resolved and error rates are back to normal.Affected components Speech to Text (Operational)
Other live feeds
Continuously updated from official provider status pages.