OpenAI compatible API · Attested · Public status
Nebius Token Factory performance
Review measured TTFT, effective throughput, uptime, and sampled model routes for Nebius Token Factory on TrustedRouter using metadata-only production probes.
Onebase URL to migrate
100sof models and routes
0prompt or output logs. Always.
Nebius Token Factorynebius
33 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 | 1417 ms |
|---|---|
| p95 TTFT | 3825 ms |
| Effective throughput | 125 tok/s n=4 |
| Uptime | 100.00% |
Measured model routes
| Model | p50 TTFT | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|
| NousResearch/Hermes-4-405B | 995 ms | — | 100.00% | — | 2 |
| reflection/beam | 1045 ms | — | 100.00% | — | 2 |
| deepseek/deepseek-v4.1-flash-1017 | 1127 ms | — | 100.00% | — | 1 |
| Qwen/Qwen3-235B-A22B-Instruct-2507 | 1128 ms | — | 100.00% | — | 2 |
| openai/gpt-oss-120b | 1164 ms | — | 100.00% | — | 1 |
| qwen/qwen3.8-27b | 1187 ms | — | 100.00% | — | 2 |
| nvidia/nemotron-3.5-super-vl | 1229 ms | — | 100.00% | — | 2 |
| openbmb/MiniCPM-V-4_5 | 1240 ms | — | 100.00% | — | 1 |
| meta-models/muse-spark-open | 1361 ms | — | 100.00% | — | 2 |
| nvidia/nemotron-3_5-lightning | 1417 ms | — | 100.00% | — | 3 |
| deepseek/deepseek-v4-flash-0731 | 1499 ms | — | 100.00% | — | 2 |
| z-ai/glm-5.3 | 1552 ms | 134 tok/s n=1 | 100.00% | — | 3 |
| zai-org/GLM-5.1 | 1761 ms | — | 100.00% | — | 2 |
| moonshotai/Kimi-K2.6 | 1987 ms | — | 100.00% | — | 1 |
| z-ai/glm-5.3-flash | 2138 ms | — | 100.00% | — | 3 |
| deepseek/deepseek-v4.1-flash | 2150 ms | 231 tok/s n=1 | 100.00% | — | 1 |
| google/gemma-3-27b-it | 2613 ms | — | 100.00% | — | 3 |
| moonshotai/kimi-k3 | — | 117 tok/s n=2 | — | — | 0 |