OpenAI compatible API · Attested · Public status
Crusoe performance
Measured TTFT, TTFB, effective throughput, uptime, and sampled model routes for Crusoe.
Onebase URL to migrate
100sof models and routes
0prompt or output logs. Always.
crusoe
209 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 | 2599 ms |
|---|---|
| p95 TTFT | 4478 ms |
| p50 TTFB | 2690 ms |
| Effective throughput | — |
| Uptime | 97.61% |
Measured model routes
| Model | p50 TTFT | p50 TTFB | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|---|
| meta-llama/llama-3.3-70b-instruct | 2360 ms | 2360 ms | — | 100.00% | — | 16 |
| deepseek/deepseek-v4-flash | 2402 ms | 2402 ms | — | 100.00% | — | 19 |
| yutori/n1.5 | 2436 ms | 2435 ms | — | 100.00% | — | 20 |
| nvidia/nemotron-3-super-120b-a12b | 2532 ms | 2532 ms | — | 100.00% | — | 12 |
| nvidia/nemotron-3-nano-30b-a3b | 2599 ms | 2597 ms | — | 100.00% | — | 15 |
| qwen/qwen3-235b-a22b-2507 | 2602 ms | 2601 ms | — | 100.00% | — | 12 |
| nvidia/nemotron-3-nano-omni-reasoning-30b-a3b | 2625 ms | 2625 ms | — | 100.00% | — | 14 |
| google/gemma-4-31b-it | 2632 ms | 2632 ms | — | 100.00% | — | 8 |
| openai/gpt-oss-120b | 3235 ms | 3235 ms | — | 100.00% | — | 12 |
| z-ai/glm-5.2 | 3267 ms | 3267 ms | — | 100.00% | — | 18 |
| z-ai/glm-5.1 | 3693 ms | 3693 ms | — | 100.00% | — | 21 |
| moonshotai/kimi-k2.6 | 2908 ms | 2908 ms | — | 93.75% | — | 16 |
| deepseek/deepseek-v3-0324 | 2585 ms | 2585 ms | — | 92.86% | — | 14 |
| deepseek/deepseek-v4-pro | 2431 ms | 2431 ms | — | 75.00% | — | 12 |