‹ Back to incident history

Scheduled maintenance: Serverless inference

ResolvedMinor incidentFireworks AI
Sunday, November 15, 2026 08:00 – Sunday, November 15, 2026 08:27 (27 min)
Affected components
Serverless / Kimi K3
Serverless / Deepseek V4.1 Flash
Serverless / GLM 5.3 Flash US
Serverless / MiniMax M3.1 Flash
Serverless / Kimi K3.1 Pro
Serverless / Muse Spark Open US
Serverless / GLM 5.3 Fast
Serverless / Qwen 3.8 Max
Serverless / GLM 5.3
Serverless / GLM 5.3 US
Serverless / Minimax M3 US
Serverless / Deepseek V4.1 Flash US
Serverless / Ember-1
Serverless / DeepSeek V4.1 Flash 1017
Serverless / DeepSeek V4.1 Flash 1017 US
Serverless / Nemotron 3.5 Lightning US
Serverless / Kimi K3 US
Serverless / Kimi K3 Fast
Serverless / MiniMax M3
Serverless / GLM 5.3 Flash
Serverless / Nemotron 3 Ultra US
Updates
maintenance_in_progressSunday, November 15, 2026 08:00

Scheduled maintenance has started. Some serverless requests may see elevated errors or latency for several minutes while this is in progress. We will update this page when it is complete.

maintenance_scheduledSaturday, November 14, 2026 10:57

We will be doing scheduled network maintenance with one of our North American infrastructure providers on Saturday, November 14, between 4:00 PM and 5:00 PM PT. The work will improve network stability. During this window, some serverless models may see a higher rate of errors (such as 503s), higher latency, or timeouts for several minutes. Where we can, we will send traffic to other regions. Requests to other models and regions should not be affected. We recommend retrying failed requests with exponential backoff. No other action is needed. We will post an update when the maintenance is complete.

Incident data from the Fireworks AI official status page:Read the official incident

About this data

Where does the incident data for Fireworks AI come from?

Data comes from the Fireworks AI official status page, syncs every 5 minutes, and covers 90 days. The detail page links to the original report.

How do I read the component status bars?

Each block is one day. Green means no issue was reported, yellow means degraded performance, red means a major outage, and gray means no data. Availability covers the last 90 days.

Why is history missing for some services?

Only services with a public official status page are included.

How is duration calculated?

Duration runs from the official start time to recovery, or to the current time for ongoing incidents. Incidents spanning multiple days appear on each affected day.

See the full service list and live status in the status overview