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
126 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 | 2834 ms |
|---|---|
| p95 TTFT | 4198 ms |
| Effective throughput | 89 tok/s n=4 |
| Uptime | 100.00% |
Measured model routes
| Model | p50 TTFT | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|
| minimax/minimax-m3 | 1034 ms | 234 tok/s n=1 | 100.00% | — | 29 |
| prism-ml/ternary-bonsai-27b | 2834 ms | — | 100.00% | — | 79 |
| meta-llama/llama-3.3-70b-instruct | 797 ms | — | 100.00% | — | 2 |
| z-ai/glm-5.2 | 1094 ms | 77 tok/s n=1 | 100.00% | — | 4 |
| qwen/qwen3.8-2.4t-a95b | 1227 ms | — | 100.00% | — | 2 |
| moonshotai/kimi-k3 | 1585 ms | 101 tok/s n=1 | 100.00% | — | 2 |
| openai/gpt-oss-120b | 1826 ms | 47 tok/s n=1 | 100.00% | — | 1 |
| thinkingmachines/inkling | 2302 ms | — | 100.00% | — | 2 |
| qwen/qwen3.5-9b | 2594 ms | — | 100.00% | — | 2 |
| z-ai/glm-5.3-flash | 2707 ms | — | 100.00% | — | 3 |