Partial Service Disruption
OutageThe incident has been resolved
Is Cerebras down?
Cerebras is up and running. Cerebras serves low-latency inference for AI models.
Last 90 days
99.77%
Estimated uptime
Estimated uptime
90 days agoToday
Worked out from the incidents Cerebras reported, so it is an estimate and not a measurement.
Cerebras reliability, last 90 days
3Incidents reported
1Full outages
1 minTypical duration
4 h 58 minLongest incident
Last incident: “Partial Service Disruption”, 28 days ago.
Components
GPT-OSS-120BOperational
Developer ConsoleOperational
Qwen-3.8-27bOperational
Recent incidents
Resolved
DegradedBetween 11:00 and 11:15 UTC users experienced partial degradation with Gemma-4-31B. Service has been restored.
Partial Service Disruption
DegradedGPT-OSS-120B was having partial service disruption between 18:20 UTC to 18:45 UTC. This is resolved now.
Partial Service Disruption
DegradedAWS billing issue is fixed now.
qwen-3-235b-a22b-instruct-2507 Service Unavailable
OutageThis incident has been resolved.
Questions
Is Cerebras down right now?
Cerebras’s official status page says: All Systems Operational. This page checks it again when you load it, at most every 5 minutes.
When was Cerebras last down?
The most recent incident Cerebras reported was “Partial Service Disruption”, 28 days ago (2026-09-25 12:16 UTC). Its last full outage started 28 days ago.
Which parts of Cerebras are tracked here?
Cerebras reports 3 components on its status page, including GPT-OSS-120B, Developer Console, Qwen-3.8-27b. Each is checked again whenever this page refreshes.
What should I do if Cerebras is down?
Retry with exponential backoff, fall back to another model or provider if your product allows it, and avoid hammering the API while it recovers.
One Cerebras model is degraded — do all models fail?
No — the status page tracks individual models (for example the hosted open-weight models) plus the Developer Console as separate rows.
Chat works but completions on one deployment fail?
Match the failing endpoint to its model row — Cerebras' page is per-model, so a single-deployment event leaves the rest of the API untouched.
How do I tell if Cerebras is down for everyone or only for me?
If the status above is operational but Cerebras still fails for you, the cause is probably on your side: try another network, clear the cache, or check your own connection. If it shows a degradation or an outage, others are affected too.
How can I get alerted when my own site goes down?
OpenUptime checks your site every minute and sends an alert to Slack, Telegram, email or a webhook the moment it stops responding. It is open source and free to start.