Maintenance ended

Maintenance is complete ahead of schedule. The Model APIs inference endpoint at api.runinfra.ai/v1 is generally available and serving normally. The website, dashboard, and the agents and optimization surfaces continue operating normally.

The Model APIs inference endpoint is deliberately paused while we finish pre-launch preparation. This is planned work, not an outage.

The gateway at api.runinfra.ai stays reachable and answers authenticated requests with the documented paused response, so clients can integrate against it now. Inference requests will not run until the endpoint returns.

Scheduled return: August 17, 2026, 9:00 AM UTC. This is the same return time published on the model page and in our machine readable listings.

The website, dashboard, and the agents and optimization surfaces are unaffected and operating normally.