Status

API status

The model runs on our own GPU system. These are operator-run probes, not a guarantee.

Status

Flonno API

Latest check passing
Thu
Fri
Sat
Sun
Mon
Tue
Wed
7-day availability

Seven-day probes

10,059 checks, 10,050 passed. Latest check: 18 November 2026 at 20:21:02 UTC. Measurements are recent.

Time to first token, p50 / p95
0.91 s / 1.45 s
Output speed, p50 / p95
115 tok/s / 127 tok/s

Operator-run synthetic inference probes. Output speed is measured from the first to the last streamed token, with thinking disabled and a 256-token output budget. Time to first token is reported separately. Not authenticated gateway availability, customer traffic, independent testing or an SLA.

Capacity

One GPU system, two generation lanes, and a shared 131,072-token KV pool. Up to 16 pending requests and a 600-second pending timeout. Per account: 60 generation requests per minute and 10 concurrent requests. The gateway timeout is 900 seconds. There is no fallback to another AI API provider.

Status JSON · Limits