Public provider healthOutages and degraded endpoints reported by official sources.
Provider status
Know when the API is down before you debug your code.
Live official-source health for the AI APIs your product depends on, plus a one-year timeline of outages, degradations, model changes, and deprecations.
Current provider signal
Tracked APIs are operational
Regional, model, and account-tier behavior can still differ.6Providers tracked
0Active incidents
152Events in timeline
15 minRefresh cadence
Current health
Provider status now
Each card links to the provider's official status source. Timeline counts reflect material incidents and platform changes recorded during the last year.
OpenAI APIOpenAIOperationalOfficial source checked9 material events in the one-year timeline
Claude APIAnthropicOperationalOfficial source checked35 material events in the one-year timeline
Voice APIElevenLabsOperationalOfficial source checked25 material events in the one-year timeline
Inference APIReplicateOperationalOfficial source checked25 material events in the one-year timeline
Speech APIDeepgramOperationalOfficial source checked50 material events in the one-year timeline
Gemini APIGoogleOperationalOfficial source checked8 material events in the one-year timelineMigration watch
Model changes that can break production
Confirm model IDs before the next deploy.
Google Gemini
Imagen 4 and Gemini 3 Image model shutdown announced
Google announced that listed Imagen 4 and Gemini 3 Image model IDs will shut down on August 17, 2026. Production integrations should migrate before that date.
Read migration sourceGoogle Gemini
Gemini 2.0 Flash model IDs shut down
Gemini 2.0 Flash and Flash-Lite stable model IDs were shut down. Google recommends moving to Gemini 3.5 Flash or Gemini 3.1 Flash-Lite.
Read migration sourceGoogle Gemini
Gemini 2.0 model shutdown announced
Google announced the June 1 shutdown of Gemini 2.0 Flash and Flash-Lite model IDs, giving production users a migration window.
Read migration sourceModel lifecycle changesDeprecations and migrations that can break a model ID or endpoint.
Your private key healthValidity, quota, permissions, and rate limits require a dashboard check.
One-year intelligence log
What changed, and when
Material API incidents, model lifecycle changes, and migrations. Every entry links to an official provider source.
Provider
Aug 31, 2026
22:27 UTC
OpenAI APIDegradation
Elevated latency in the Responses API
All impacted services have now fully recovered.
Aug 31, 2026
08:10 UTC
Voice APIDegradation
Customer support ticketing system not receiving new tickets and ticket replies
We’ll mark this incident as resolved. Our ticketing system is operational again, though the outage materially affected its setup. We’re working with the team to restore full functionality, but capacity is sufficient to consider the inciden…
Aug 28, 2026
17:22 UTC
Claude APIOutage
Elevated errors on Claude Code and Claude Cowork
The issue affecting Claude Cowork and Claude Code on the web has been resolved.
Aug 27, 2026
13:05 UTC
Voice APIDegradation
Batch calling degraded in EU residency
Latency has since decreased, and we are working on adding more alerts on our side to help us identify and respond to similar latency issues earlier.
Aug 26, 2026
21:17 UTC
Inference APIOutage
A100 hardware partial outage
We've been below capacity for A100 hardware for the past 30 minutes and all queues have been drained. Thanks for your patience 💖
Aug 26, 2026
17:16 UTC
Voice APIDegradation
Elevated latency on Text-to-Speech and Speech-to-Text Requests
The issue has been resolved. A root cause analysis will be shared shortly.
Aug 25, 2026
18:01 UTC
Speech APIDegradation
[Deepgram EU API] Flux-Multi STT Outage
Resolved Between 07:13 and 09:00 UTC, flux-general-multi Speech-to-Text requests on api.eu.deepgram.com failed shortly after the WebSocket connection was established, returning an internal server error on the first audio and closing the co…
Aug 25, 2026
16:41 UTC
Speech APIDegradation
[Deepgram API] Elevated Streaming STT Error Rates
Between 23:18 and 23:45 UTC, a portion of streaming STT connection requests failed with 5XX errors and timeouts. While rolling out a new data center, we detected streaming errors and immediately removed it from rotation. Traffic to the new…
Aug 24, 2026
08:49 UTC
Voice APIDegradation
Media not loading across several products
Our engineers have traced media loading error rates as they have returned to baseline. Studio, Audiobooks, Image & Video, Music and Dubbing products should now functional normally.
Aug 24, 2026
05:06 UTC
Claude APIOutage
Elevated errors for multiple models
This issue has been resolved. From 9:50pm PT / 04:50 UTC through 00:36am PT / 07:36 UTC, users saw elevated errors on requests to Claude models, including Claude Opus 5 and Fable 5.
Aug 20, 2026
19:16 UTC
Claude APIOutage
Elevated errors on requests to multiple models
This issue has been resolved.
Aug 20, 2026
18:32 UTC
Claude APIDegradation
Elevated errors on Google connectors
This issue has been resolved.
Aug 20, 2026
16:58 UTC
OpenAI APIDegradation
Elevated Codex API authentication errors
All impacted services have now fully recovered.
Aug 20, 2026
15:52 UTC
Speech APIDegradation
Elevated 5XX Error Rate for v1/auth/grant
This incident has been resolved.
Aug 19, 2026
09:42 UTC
Claude APIDegradation
Degraded performance for Claude Opus 5 and Claude Haiku 4.5
This incident has been resolved.
Aug 18, 2026
16:20 UTC
Claude APIDegradation
Degraded performance for multiple models
The issue affecting Claude Opus 5 has been resolved. Impact occurred from 16:11 to 18:23 UTC.
Aug 18, 2026
12:39 UTC
Voice APIDegradation
Inbound Twilio calls experiencing intermittent connection failures
Inbound calls are currently operational.
Aug 17, 2026
13:56 UTC
Claude APIDegradation
Degraded performance for Claude Opus 5, Claude Sonnet 5
This incident has been resolved.
Provider operational, but your requests still fail?
Public status shows broad provider incidents. Invalid credentials, exhausted quota, project restrictions, revoked keys, and account-level rate limits remain private and appear inside your API Key Health dashboard.
Provider status FAQ
Clear answers for AI API outage and quota questions
What is the API Key Health provider status page?
The API Key Health provider status page is a public timeline for major AI API outages, degraded API endpoints, model launches, model shutdowns, and migration events across providers such as OpenAI, Google Gemini, Anthropic Claude, ElevenLabs, Replicate, and Deepgram.
Does provider status prove that my API key is working?
No. Public provider status only shows broad incidents. A provider can be operational while your own key is invalid, revoked, rate-limited, out of quota, missing permissions, or blocked by a project restriction. API Key Health checks those private key-level issues inside your dashboard.
How often does this AI API status page refresh?
The public status snapshot refreshes every 15 minutes. It combines live official provider feeds with a curated one-year timeline of material incidents and model lifecycle changes.
Which API providers are tracked?
The page tracks OpenAI API, Google Gemini API, Anthropic Claude API, ElevenLabs Voice API, Replicate Inference API, and Deepgram Speech API. API Key Health also supports private API key monitoring for many additional providers inside the product.
Where does the status data come from?
Live incident signals come from official provider status feeds and official provider health pages. Curated model lifecycle entries link to official changelogs, status pages, or provider release notes.