OpenAI compatible API · Attested · Public status
Baseten performance
Review measured TTFT, effective throughput, uptime, and sampled model routes for Baseten on TrustedRouter using metadata-only production probes.
Onebase URL to migrate
100sof models and routes
0prompt or output logs. Always.
Basetenbaseten
493 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 | 2190 ms |
|---|---|
| p95 TTFT | 4796 ms |
| Effective throughput | 254 tok/s n=2 |
| Uptime | 99.59% |
Measured model routes
| Model | p50 TTFT | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|
| deepseek/deepseek-v4-flash-0731 | 2190 ms | — | 99.62% | — | 260 |
| z-ai/glm-5.3-fast | 850 ms | — | 100.00% | — | 3 |
| moonshotai/kimi-k2.7-code | 958 ms | 356 tok/s n=1 | 100.00% | — | 7 |
| z-ai/glm-5.3 | 1321 ms | — | 100.00% | — | 2 |
| thinkingmachines/inkling-1m | 1438 ms | — | 100.00% | — | 1 |
| nvidia/nemotron-3-ultra-550b-a55b | 1532 ms | — | 100.00% | — | 1 |
| deepseek/deepseek-v4-pro-0813 | 1743 ms | — | 99.24% | — | 132 |
| moonshotai/kimi-k2.6 | 1813 ms | — | 100.00% | — | 2 |
| deepseek/deepseek-v4.1-flash | 4055 ms | — | 100.00% | — | 1 |
| z-ai/glm-5.2 | 4944 ms | — | 100.00% | — | 48 |
| moonshotai/kimi-k3 | — | — | 100.00% | — | 36 |
| openai/gpt-oss-120b | — | 152 tok/s n=1 | — | — | 0 |