Status

Live probe against https://microquickjs.com/v1, model Kimi-K3, every 10 minutes from a GitHub Actions runner in the United States. One request of about 1,000 input tokens and 200 output tokens. Updated ….

Availability 24h

…

Availability 7d

…

Time to first token

…p50, 24h

Output speed

…tok/s p50, 24h

Last 24 hours

Each bar is one probe. Green: succeeded. Red: failed or timed out. Height: time to first token.

24h agonow

What this measures

Availability: share of probes that returned a streamed completion with at least one token. It is the same path your requests take; no synthetic health endpoint.
Time to first token: from request sent to first content token, including the model's thinking phase, as seen from a United States vantage point. Your numbers will differ by region.
Output speed: tokens per second after the first token, single stream.

Probe traffic is excluded from every usage statistic we publish. This page carries no uptime commitment; committed and core plans get one in writing.