OpenAI compatible API · Attested · Public status
Cerebras performance
Review measured TTFT, TTFB, effective throughput, uptime, and sampled model routes for Cerebras on TrustedRouter using metadata-only production probes.
Onebase URL to migrate
100sof models and routes
0prompt or output logs. Always.
Cerebrascerebras
55 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 | 1013 ms |
|---|---|
| p95 TTFT | 6456 ms |
| p50 TTFB | 1426 ms |
| Effective throughput | 151 tok/s n=6 |
| Uptime | 89.09% |
Measured model routes
| Model | p50 TTFT | p50 TTFB | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|---|
| openai/gpt-oss-120b | 1013 ms | 1013 ms | 151 tok/s n=2 | 100.00% | — | 6 |
| cerebras/gpt-oss-120b | 1379 ms | 1379 ms | 431 tok/s n=2 | 100.00% | — | 12 |
| google/gemma-4-31b-it | 1859 ms | 1859 ms | 27 tok/s n=1 | 88.89% | — | 9 |
| cerebras/zai-glm-4.7 | 970 ms | 970 ms | 38 tok/s n=1 | 83.33% | — | 12 |
| z-ai/glm-4.7 | 951 ms | 951 ms | — | 81.82% | — | 11 |
| cerebras/gemma-4-31b | 2116 ms | 2116 ms | — | 80.00% | — | 5 |