OpenAI compatible API · Attested · Public status
DeepInfra performance
Review measured TTFT, 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
495 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 | 3121 ms |
|---|---|
| p95 TTFT | 6937 ms |
| Effective throughput | 72 tok/s n=3 |
| Uptime | 99.19% |
Measured model routes
| Model | p50 TTFT | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|
| deepseek/deepseek-v4-flash-0731 | 3121 ms | — | 99.52% | — | 418 |
| meta-models/muse-glimmer-30b | 860 ms | — | 100.00% | — | 1 |
| qwen/qwen3.7-max | 3762 ms | — | 100.00% | — | 1 |
| qwen/qwen3.5-27b | 3995 ms | — | 100.00% | — | 1 |
| nvidia/nemotron-content-safety-3.5 | 4240 ms | — | 100.00% | — | 1 |
| stepfun-ai/step-3.7-flash | 4614 ms | — | 100.00% | — | 1 |
| meta-llama/meta-llama-3.1-8b-instruct-turbo | — | — | 100.00% | — | 1 |
| mistralai/mistral-nemo-instruct-2407 | — | — | 100.00% | — | 1 |
| xiaomi/mimo-v2.5-pro | — | 23 tok/s n=1 | 100.00% | — | 68 |
| deepseek/deepseek-v4-pro-0423 | — | 152 tok/s n=1 | — | — | 0 |
| google/gemma-4-31b-it | — | 72 tok/s n=1 | — | — | 0 |
| openai/gpt-oss-120b-ultra | — | — | 0.00% | — | 1 |
| z-ai/glm-5.3 | — | — | 0.00% | — | 1 |