OpenAI compatible API · Attested · Public status
DeepInfra performance
Review measured TTFT, TTFB, effective throughput, uptime, and sampled model routes for DeepInfra on TrustedRouter using metadata-only production probes.
Onebase URL to migrate
100sof models and routes
0prompt or output logs. Always.
DeepInfradeepinfra
93 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 | 3743 ms |
|---|---|
| p95 TTFT | 10206 ms |
| p50 TTFB | 3416 ms |
| Effective throughput | 27 tok/s n=21 |
| Uptime | 98.92% |
Measured model routes
| Model | p50 TTFT | p50 TTFB | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|---|
| deepseek/deepseek-v4-flash-0731 | 6216 ms | 1904 ms | 25 tok/s n=1 | 100.00% | — | 14 |
| google/gemma-3-27b-it | 677 ms | 677 ms | — | 100.00% | — | 2 |
| qwen/qwen3-vl-30b-a3b-instruct | 761 ms | 761 ms | — | 100.00% | — | 2 |
| nvidia/nemotron-3-nano-30b-a3b | 916 ms | 915 ms | — | 100.00% | — | 3 |
| nvidia/nemotron-3.5-lightning | 941 ms | 941 ms | 82 tok/s n=1 | 100.00% | — | 2 |
| mistralai/mistral-small-24b-instruct-2501 | 976 ms | 976 ms | — | 100.00% | — | 2 |
| sao10k/l3-8b-lunaris-v1-turbo | 1190 ms | 1190 ms | — | 100.00% | — | 1 |
| bytedance/seed-2.0-mini | 1414 ms | 1414 ms | — | 100.00% | — | 2 |
| moonshotai/kimi-k3 | 1464 ms | 1464 ms | 16 tok/s n=1 | 100.00% | — | 1 |
| qwen/qwen3-next-80b-a3b-instruct | 1500 ms | 1500 ms | — | 100.00% | — | 2 |
| google/gemma-4-26b-a4b-it | 1669 ms | 1668 ms | — | 100.00% | — | 1 |
| deepseek/deepseek-r1-0528 | 2459 ms | 2459 ms | — | 100.00% | — | 4 |
| google/gemma-4-31b-it | 2482 ms | 2481 ms | 16 tok/s n=1 | 100.00% | — | 1 |
| google/gemma-4-31b-it-turbo | 3013 ms | 3013 ms | — | 100.00% | — | 2 |
| minimax/minimax-m3 | 3067 ms | 3067 ms | 17 tok/s n=2 | 100.00% | — | 4 |
| z-ai/glm-4.6 | 3417 ms | 3416 ms | — | 100.00% | — | 3 |
| meta-llama/llama-4-scout-17b-16e-instruct | 3743 ms | 3743 ms | — | 100.00% | — | 1 |
| deepseek/deepseek-v3.2 | 3842 ms | 3842 ms | — | 100.00% | — | 1 |
| qwen/qwen3.5-35b-a3b | 3844 ms | 3844 ms | — | 100.00% | — | 1 |
| google/gemma-4-e4b-it | 3923 ms | 3923 ms | — | 100.00% | — | 1 |
| z-ai/glm-4.7 | 4064 ms | 4064 ms | — | 100.00% | — | 1 |
| meta-llama/llama-guard-4-12b | 4080 ms | 4079 ms | — | 100.00% | — | 1 |
| meta-llama/llama-3.1-70b-instruct | 4103 ms | — | — | 100.00% | — | 3 |
| microsoft/phi-4 | 4138 ms | 4138 ms | — | 100.00% | — | 2 |
| minimax/minimax-m2.7 | 4209 ms | 4209 ms | — | 100.00% | — | 2 |
| qwen/qwen3.6-35b-a3b | 4232 ms | 4231 ms | — | 100.00% | — | 2 |
| qwen/qwen3.5-397b-a17b | 4261 ms | 4261 ms | — | 100.00% | — | 1 |
| qwen/qwen3-vl-235b-a22b-instruct | 4319 ms | 4319 ms | — | 100.00% | — | 1 |
| thinkingmachines/inkling-small | 4363 ms | 4363 ms | 27 tok/s n=1 | 100.00% | — | 1 |
| qwen/qwen2.5-72b-instruct | 4582 ms | 4582 ms | — | 100.00% | — | 1 |
| bytedance/seed-2.0-code | 4631 ms | 4630 ms | — | 100.00% | — | 1 |
| google/gemma-4-31b-it-ultra | 5622 ms | 5621 ms | — | 100.00% | — | 1 |
| qwen/qwen3.7-max | 6244 ms | 6244 ms | 77 tok/s n=1 | 100.00% | — | 1 |
| google/gemini-3.1-pro | 6262 ms | 6262 ms | — | 100.00% | — | 1 |
| google/gemini-3.7-flash | 7103 ms | 7103 ms | 51 tok/s n=1 | 100.00% | — | 1 |
| bytedance/seed-2.0-pro | — | — | — | 100.00% | — | 2 |
| gryphe/mythomax-l2-13b | — | — | — | 100.00% | — | 1 |
| meta-models/muse-glimmer-30b | — | — | — | 100.00% | — | 1 |
| mistralai/mistral-nemo-instruct-2407 | — | — | — | 100.00% | — | 2 |
| nousresearch/hermes-3-llama-3.1-405b | — | — | — | 100.00% | — | 2 |
| nvidia/nemotron-3-ultra-550b-a55b | — | — | 15 tok/s n=1 | 100.00% | — | 2 |
| openai/gpt-oss-120b | — | — | 40 tok/s n=1 | 100.00% | — | 2 |
| qwen/qwen3-max | — | — | — | 100.00% | — | 1 |
| qwen/qwen3-max-thinking | — | — | — | 100.00% | — | 1 |
| stepfun-ai/step-3.7-flash | — | — | — | 100.00% | — | 1 |
| tencent/hy3 | — | — | — | 100.00% | — | 2 |
| thinkingmachines/inkling | — | — | 10 tok/s n=1 | 100.00% | — | 2 |
| qwen/qwen3-235b-a22b-thinking-2507 | 3055 ms | 3054 ms | — | 75.00% | — | 4 |
| deepseek/deepseek-v4-flash | — | — | 39 tok/s n=2 | — | — | 0 |
| deepseek/deepseek-v4-pro-0423 | — | — | 19 tok/s n=2 | — | — | 0 |
| inclusionai/ling-3.0-flash | — | — | 55 tok/s n=1 | — | — | 0 |
| moonshotai/kimi-k2.6 | — | — | 22 tok/s n=1 | — | — | 0 |
| qwen/qwen3.8-2.4t-a95b | — | — | 39 tok/s n=1 | — | — | 0 |
| z-ai/glm-5.2 | — | — | 52 tok/s n=2 | — | — | 0 |