Gonka Network
Network Status
Real-time model availability and latency. Automatically updated every minute.
Overall
All systems operational
No data
Incidents in the last 48 hours
No incidents recorded in the last 48 hours.
Frequently asked questions
- What does 'Quality degradation' mean?
- Some test requests failed or had a noticeable delay. The service is still working, but users may see timeouts or slowdowns. This is usually a temporary peak load on the network.
- What does 'Failure' mean?
- More than half of the requests do not receive a response within 15 seconds. The model is effectively unavailable. We open an incident and continue checks — the status is automatically restored as soon as the network responds.
- Where does the data come from?
- Our backend makes a signed inference request directly to Gonka network nodes once a minute via the gonka-openai SDK. We do not depend on third-party monitors — exactly the same nodes as for user requests.
- How often is the page updated?
- The widget automatically reloads data every minute. To keep the data fresh in real time, you can simply leave the tab open.
How we measure
Once a minute, we send a test request to each Gonka network model and record the response and latency. Status is determined based on the last 15 minutes: "Outage" - if more than half of the probes failed; "Degraded performance" - if more than 20% failed, or if the response latency is more than 2.5 times the typical level.