OpenAI compatible API · Attested · Public status
Together performance
Review measured TTFT, effective throughput, uptime, and sampled model routes for Together on TrustedRouter using metadata-only production probes.
Onebase URL to migrate
100sof models and routes
0prompt or output logs. Always.
Togethertogether
58 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 | 874 ms |
|---|---|
| p95 TTFT | 2475 ms |
| Effective throughput | 139 tok/s n=3 |
| Uptime | 84.48% |
Measured model routes
| Model | p50 TTFT | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|
| prism-ml/ternary-bonsai-27b | 874 ms | — | 96.88% | — | 32 |
| minimax/minimax-m3.1-flash | 594 ms | — | 100.00% | — | 1 |
| z-ai/glm-5.2 | 748 ms | 290 tok/s n=1 | 100.00% | — | 2 |
| minimax/minimax-m3 | 946 ms | 139 tok/s n=1 | 100.00% | — | 1 |
| openai/gpt-oss-120b | 1036 ms | — | 100.00% | — | 2 |
| z-ai/glm-5.3-flash | 1054 ms | — | 100.00% | — | 3 |
| qwen/qwen3.5-9b | 1073 ms | — | 100.00% | — | 3 |
| moonshotai/kimi-k3 | 1524 ms | 34 tok/s n=1 | 100.00% | — | 4 |
| meta-models/muse-glimmer-30b | 1652 ms | — | 100.00% | — | 1 |
| thinkingmachines/inkling | 2475 ms | — | 11.11% | 14 provider_error |
9 |
| deepseek/deepseek-v4.1-flash | — | — | — | 1 provider_error |
0 |
| qwen/qwen3.8-2.4t-a95b | — | — | — | 2 provider_error |
0 |