UtteraUttera

Service status

The API is responding

The automatic check against api.uttera.ai/health came back fine.

This check runs at most once a minute and looks at the same thing you can look at: whether api.uttera.ai/health answers. It is not an internal monitoring dashboard, and we say so you know exactly what it is worth.

100% self-supplied. All the energy the site spent over the last 30 days it generated itself: sun and batteries.

Measured at the utility company's meter, in net balance. How it is computed and what that number does not say →

Availability

ServiceLast 24 hLast 30 days
Transcription100 %100 %
Speech100 %100 %
Cloned voice100 %100 %
Translation100 %100 %
Summaries100 %100 %
Voice analysis100 %100 %
About the 30-day column: the service opened on 15 September 2026, 52 days ago, so that window includes time before launch, when the system was being tested and restarted on purpose. Don't take it as a reference yet: it won't be comparable until there are 30 full days of service.

Measured every minute: a service counts as available when at least one node in its group is serving, which is what a real request experiences. Measuring per node would give a worse and less true number — one node down with two others up is not an outage.

Incident history

What we report here

We publish every interruption that affected customer requests, with its duration, which services were affected and what caused it. Including the short ones: a history that only shows the big outages is no use for judging a provider.

The availability commitment depends on your plan and is in the documentation. The free, startup and developer plans carry no SLA.

If something is failing for you right now

The first thing is knowing whether it is yours or ours. GET https://api.uttera.ai/health answers without a key. And every API response carries an X-Request-Id header: keep it when something fails, because with that identifier we can tell you exactly what happened to your request. Write to support@uttera.ai.