OpenAI compatible API · Attested · Public status
Crusoe performance
Review measured TTFT, TTFB, effective throughput, uptime, and sampled model routes for Crusoe on TrustedRouter using metadata-only production probes.
Onebase URL to migrate
100sof models and routes
0prompt or output logs. Always.
Crusoecrusoe
64 samples
Continuously sampled provider performance. TrustedRouter reports unsupported route and probe-configuration rows separately from provider downtime. Prompt and output content is not stored.
| p50 TTFT | 1659 ms |
|---|---|
| p95 TTFT | 5106 ms |
| p50 TTFB | 1958 ms |
| Effective throughput | 38 tok/s n=6 |
| Uptime | 100.00% |
Measured model routes
| Model | p50 TTFT | p50 TTFB | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|---|
| google/gemma-4-31b-it | 952 ms | 952 ms | 26 tok/s n=1 | 100.00% | — | 6 |
| nvidia/nemotron-3-super-120b-a12b | 1128 ms | 1128 ms | — | 100.00% | — | 4 |
| yutori/n1.5 | 1258 ms | 1258 ms | — | 100.00% | — | 6 |
| deepseek/deepseek-v4-flash | 1461 ms | 1460 ms | 35 tok/s n=2 | 100.00% | — | 4 |
| meta-llama/llama-3.3-70b-instruct | 1474 ms | 1474 ms | — | 100.00% | — | 6 |
| deepseek/deepseek-v3-0324 | 1608 ms | 1608 ms | — | 100.00% | — | 2 |
| nvidia/nemotron-3-nano-omni-reasoning-30b-a3b | 1658 ms | 1657 ms | — | 100.00% | — | 3 |
| nvidia/nemotron-3-nano-30b-a3b | 1659 ms | 1659 ms | — | 100.00% | — | 5 |
| deepseek/deepseek-v4-pro | 1815 ms | 1815 ms | 42 tok/s n=1 | 100.00% | — | 7 |
| openai/gpt-oss-120b | 1958 ms | 1958 ms | 120 tok/s n=2 | 100.00% | — | 4 |
| z-ai/glm-5.1 | 2522 ms | 2522 ms | — | 100.00% | — | 4 |
| qwen/qwen3-235b-a22b-2507 | 3348 ms | 3348 ms | — | 100.00% | — | 6 |
| moonshotai/kimi-k2.6 | 3423 ms | 3423 ms | — | 100.00% | — | 3 |
| z-ai/glm-5.2 | 3908 ms | 3908 ms | — | 100.00% | — | 4 |