Status
Checking status…
Availability is measured from real requests through the gateway. Each day is one cell: 99.5 % or more of requests succeeded, between 95 % and 99.5 %, below 95 %, or no traffic that day. First-token latency is the daily median. A model without enough recent traffic to measure shows "Unverified metrics" rather than a state we have not observed. Incidents are posted here and in the console.
Refreshes every minute
Components
Models · last 30 days
≥ 99.5 %95–99.5 %< 95 %No traffic
ModelNowAvailability by day30 dTTFT p50tok/s
Select a day (click, tap or use the arrow keys) to read its figures.