Use cases
Software Products E-commerce MSPs Schools Development & Marketing DevOps Agencies Help Desk
Company
Internet Status Blog Pricing Log in Get started free

Cerebras Outage History

Every past Cerebras outage tracked by IsDown, with detection times, duration, and resolution details.

There were 31 Cerebras outages since January 2025. The 24 outages from the last 12 months are summarized below, with incident details, duration, and resolution information.

Major May 1, 2026

May 2026: qwen-3-235b-a22b-instruct-2507 Service Unavailable

Detected May 1, 2026 7:59 AM UTC · Resolved May 1, 2026 8:59 AM UTC · Duration about 1 hour

The Cerebras qwen-3-235b-a22b-instruct-2507 service became completely inaccessible, preventing users from accessing the AI model. The major outage lasted approximately 1 hour before service was fully restored. The incident was resolved without any detailed explanation of the root cause.

Major March 17, 2026

March 2026: Z.ai GLM 4.7 Service Unavailable

Detected Mar 17, 2026 12:21 PM UTC · Resolved Mar 17, 2026 6:33 PM UTC · Duration about 6 hours

Cerebras Z.ai GLM 4.7 service became completely inaccessible, preventing users from utilizing the service for 6.2 hours. The incident was classified as major due to the total service unavailability. The service was successfully restored and the incident was resolved.

Major March 17, 2026

March 2026: Z.ai GLM 4.7 Service Unavailable

Detected Mar 17, 2026 11:14 AM UTC · Resolved Mar 17, 2026 12:11 PM UTC · Duration about 1 hour

The Cerebras Z.ai GLM 4.7 service became completely inaccessible, preventing users from accessing the platform. The outage lasted 57 minutes before the team deployed a fix that fully restored service capabilities.

Minor March 6, 2026

March 2026: Partial Service Disruption

Detected Mar 6, 2026 5:29 AM UTC · Resolved Mar 6, 2026 5:54 AM UTC · Duration 25 minutes

The Cerebras Qwen 235B endpoint experienced a partial service disruption for 45 minutes (4:53-5:38 UTC) due to a transient network issue. The disruption affected the availability and performance of the Qwen 235B model endpoint. The network issue was identified and resolved, restoring normal service operations.