Status operational
Updated Wed, 28 Oct 2026 00:30:21 GMT. Workers ready: 1/1.
| Metric | Last 24 hours | Last 7 days |
|---|---|---|
| Uptime | 100.00% | 100.00% |
| Requests | 519 | 519 |
| Error rate | 0.00% | 0.00% |
| Time to first token, p50 | 339 ms | 339 ms |
| Time to first token, p95 | 442 ms | 442 ms |
| Output speed, p50 | 94.4 tok/s | 94.4 tok/s |
Serving configuration
| Model | Deployment | Quantization | Engine | Hardware | Context |
|---|---|---|---|---|---|
qwen3.8-27b-uncensored | q38-dev-mock | nvfp4 | mock | mock engine (no GPU) | 131,072 |
Uptime is the share of one-minute control-plane probes that found a ready worker. Error rate counts engine failures and invalid outputs; client cancellations and capacity rejections are excluded.