Nebius Token Factory
Explore Nebius Token Factory models on TrustedRouter with current routes, token pricing, policy sources, privacy posture, regional availability, and API support.
Nebius Token Factorynebius
No logs
| Provider | Nebius Token Factory |
|---|---|
| Provider website | https://nebius.com/ai-studio |
| Models | 29 public models |
| Prepaid routes | 27 |
| BYOK routes | 29 |
| Zero data retention | yes |
| Confidential compute | not claimed |
| Provider E2EE | not claimed |
| Policy note | Marked ZDR via TrustedRouter's arrangement — Nebius RETAINS inputs/outputs by default (for speculative decoding); zero retention is an opt-in control, which the deployed Nebius account has enabled. Nebius does not train on customer data. Policy source |
Measured performance
111 samplesContinuously sampled across Nebius Token Factory's routed models: p50 TTFT, effective throughput, and success rate. Effective throughput uses provider-reported output tokens over complete request time. Unsupported route and probe-configuration rows are separated from provider downtime. No prompt or output content stored.
| p50 TTFT | 1953 ms |
|---|---|
| Effective throughput | 63 tok/s n=7 |
| Uptime | 97.30% |
| Model | p50 TTFT | p50 TTFB | Effective throughput | Uptime | Config excluded | Availability samples |
|---|---|---|---|---|---|---|
| NousResearch/Hermes-4-405B | 948 ms | — | — | 100.00% | — | 6 |
| Qwen/Qwen3-32B | 1019 ms | 1019 ms | — | 100.00% | — | 7 |
| openbmb/MiniCPM-V-4_5 | 1107 ms | 1309 ms | — | 100.00% | — | 2 |
| nvidia/Nemotron-3-Nano-Omni | 1178 ms | — | — | 100.00% | — | 1 |
| nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 | 1185 ms | 1185 ms | — | 100.00% | — | 3 |
| nvidia/nemotron-3-super-120b-a12b | 1209 ms | 2243 ms | — | 100.00% | — | 2 |
| MiniMaxAI/MiniMax-M3 | 1338 ms | 1337 ms | 54 tok/s n=1 | 100.00% | — | 7 |
| nvidia/nemotron-3-ultra-550b-a55b | 1657 ms | 1657 ms | 140 tok/s n=2 | 100.00% | — | 2 |
| moonshotai/Kimi-K2.7-Code | 1768 ms | 4574 ms | — | 100.00% | — | 2 |
| nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B | 1812 ms | 1812 ms | — | 100.00% | — | 5 |
| moonshotai/Kimi-K2.6 | 1865 ms | — | — | 100.00% | — | 1 |
| Qwen/Qwen2.5-VL-72B-Instruct | 1935 ms | 1935 ms | — | 100.00% | — | 8 |
| zai-org/GLM-5.2 | 1942 ms | 1223 ms | — | 100.00% | — | 3 |
| google/gemma-3-27b-it | 2218 ms | 2218 ms | — | 100.00% | — | 3 |
| Qwen/Qwen3-30B-A3B-Instruct-2507 | 2264 ms | 2264 ms | — | 100.00% | — | 8 |
| Qwen/Qwen3-Next-80B-A3B-Thinking | 2430 ms | 5513 ms | — | 100.00% | — | 9 |
| MiniMaxAI/MiniMax-M2.5 | 2471 ms | — | — | 100.00% | 1 probe_config_error |
6 |
| NousResearch/Hermes-4-70B | 2710 ms | 4483 ms | — | 100.00% | — | 9 |
| nvidia/Cosmos3-Super-Reasoner | 2727 ms | 2727 ms | — | 100.00% | — | 3 |
| openai/gpt-oss-120b | 2868 ms | 2868 ms | 182 tok/s n=1 | 100.00% | — | 1 |
| moonshotai/kimi-k3 | 4070 ms | 4070 ms | 63 tok/s n=2 | 100.00% | — | 1 |
| Qwen/Qwen3-235B-A22B-Instruct-2507 | 4568 ms | — | — | 100.00% | — | 6 |
| zai-org/GLM-5.1 | 6915 ms | 6915 ms | — | 100.00% | — | 3 |
| Qwen/Qwen3.5-397B-A17B | 2121 ms | — | — | 83.33% | — | 6 |
| deepseek-ai/DeepSeek-V4-Pro | 1953 ms | 1637 ms | — | 75.00% | — | 4 |
| meta-llama/Llama-3.3-70B-Instruct | 1188 ms | 1296 ms | — | 66.67% | — | 3 |
| deepseek/deepseek-v4-flash | — | — | 48 tok/s n=1 | — | — | 0 |
Nebius Token Factory performance history · Full provider & model leaderboard.
Models served by Nebius Token Factory.
Each row links to pricing, provider, benchmark, and API pages for the model.
| Model | AI IQ | Context | Endpoints | Prompt | Completion | Routes |
|---|---|---|---|---|---|---|
MiniMaxAI/MiniMax-M2.5MiniMax M2.5 |
IQ 105#56 | 204,800 | 2 | $0.315/1M | $1.26/1M | prepaid BYOK |
MiniMaxAI/MiniMax-M3MiniMax-M3 |
IQ 114#33 | 1,048,576 | 2 | $0.315/1M | $1.26/1M | prepaid BYOK |
NousResearch/Hermes-4-405BHermes 4 405B |
— | 131,072 | 2 | $1.05/1M | $3.15/1M | prepaid BYOK |
NousResearch/Hermes-4-70BHermes 4 70B |
— | 131,072 | 2 | $0.1365/1M | $0.42/1M | prepaid BYOK |
Qwen/Qwen2.5-VL-72B-InstructQwen2.5 VL 72B Instruct |
— | 32,768 | 2 | $0.2625/1M | $0.7875/1M | prepaid BYOK |
Qwen/Qwen3-235B-A22B-Instruct-2507Qwen3 235B A22B Instruct 2507 |
— | 131,072 | 2 | $0.21/1M | $0.63/1M | prepaid BYOK |
Qwen/Qwen3-30B-A3B-Instruct-2507Qwen3 30B A3B Instruct 2507 |
— | 131,072 | 2 | $0.105/1M | $0.315/1M | prepaid BYOK |
Qwen/Qwen3-32BQwen3 32B |
— | 131,072 | 2 | $0.105/1M | $0.315/1M | prepaid BYOK |
Qwen/Qwen3-Next-80B-A3B-ThinkingQwen3 Next 80B A3B Thinking |
— | 131,072 | 2 | $0.1575/1M | $1.26/1M | prepaid BYOK |
Qwen/Qwen3.5-397B-A17BQwen3.5 397B A17B |
— | 262,144 | 2 | $0.63/1M | $3.78/1M | prepaid BYOK |
deepseek-ai/DeepSeek-V4-ProDeepSeek V4 Pro |
IQ 117#25 | 1,048,576 | 2 | $1.8375/1M | $3.675/1M | prepaid BYOK |
deepseek/deepseek-v4-flashDeepSeek: DeepSeek V4 Flash 0423 |
IQ 113#35 | 1,048,576 | 2 | $0.147/1M | $0.294/1M | prepaid BYOK |
google/gemma-2-2b-itgemma 2 2b it |
— | 8,192 | 1 | $0.021/1M | $0.063/1M | BYOK |
google/gemma-3-27b-itGoogle: Gemma 3 27B |
— | 262,144 | 2 | $0.105/1M | $0.315/1M | prepaid BYOK |
meta-llama/Llama-3.3-70B-InstructLlama 3.3 70B Instruct |
— | 131,072 | 2 | $0.1365/1M | $0.42/1M | prepaid BYOK |
meta-llama/Meta-Llama-3.1-8B-InstructMeta Llama 3.1 8B Instruct |
— | 128,000 | 1 | $0.021/1M | $0.063/1M | BYOK |
moonshotai/Kimi-K2.6Kimi-K2.6 |
IQ 119#19 | 8,000 | 2 | $0.9975/1M | $4.2/1M | prepaid BYOK |
moonshotai/Kimi-K2.7-CodeKimi-K2.7-Code |
IQ 118#22 | 262,144 | 2 | $0.9975/1M | $4.2/1M | prepaid BYOK |
moonshotai/kimi-k3MoonshotAI: Kimi K3 |
IQ 123#14 | 1,048,576 | 2 | $3.15/1M | $15.75/1M | prepaid BYOK |
nvidia/Cosmos3-Super-ReasonerCosmos3-Super-Reasoner |
— | 8,000 | 2 | $0.105/1M | $0.315/1M | prepaid BYOK |
nvidia/Llama-3_1-Nemotron-Ultra-253B-v1Llama 3_1 Nemotron Ultra 253B v1 |
— | 128,000 | 2 | $0.63/1M | $1.89/1M | prepaid BYOK |
nvidia/NVIDIA-Nemotron-3-Nano-30B-A3BNVIDIA Nemotron 3 Nano 30B A3B |
— | 131,072 | 2 | $0.063/1M | $0.252/1M | prepaid BYOK |
nvidia/Nemotron-3-Nano-OmniNemotron 3 Nano Omni |
— | 131,072 | 2 | $0.063/1M | $0.252/1M | prepaid BYOK |
nvidia/nemotron-3-super-120b-a12bNVIDIA: Nemotron 3 Super |
— | 1,000,000 | 2 | $0.315/1M | $0.945/1M | prepaid BYOK |
nvidia/nemotron-3-ultra-550b-a55bNVIDIA: Nemotron 3 Ultra |
— | 512,288 | 2 | $1.05/1M | $3.15/1M | prepaid BYOK |
openai/gpt-oss-120bOpenAI: gpt-oss-120b |
IQ 105#53 | 131,072 | 2 | $0.1575/1M | $0.63/1M | prepaid BYOK |
openbmb/MiniCPM-V-4_5openbmb/MiniCPM-V-4_5 |
— | 8,000 | 2 | $0.6909/1M | $1.1655/1M | prepaid BYOK |
zai-org/GLM-5.1GLM 5.1 |
IQ 114#32 | 204,800 | 2 | $1.47/1M | $4.62/1M | prepaid BYOK |
zai-org/GLM-5.2GLM-5.2 |
IQ 120#16 | 1,048,576 | 2 | $1.47/1M | $4.62/1M | prepaid BYOK |
Questions
Does Nebius Token Factory have zero data retention?
TrustedRouter records Nebius Token Factory as supporting provider-level zero data retention based on the policy source linked on this page. This is a provider policy claim, separate from TrustedRouter's content-stateless real-time gateway and from end-to-end confidential compute.
Is Nebius Token Factory end-to-end encrypted?
TrustedRouter does not currently mark Nebius Token Factory as end-to-end encrypted at the provider boundary. The TrustedRouter gateway is still attested, but the selected provider normally receives the request in order to run the model. Use trustedrouter/e2e for the stronger route requirement.
Which Nebius Token Factory models are available through TrustedRouter?
This page currently lists 29 public Nebius Token Factory models, with live pricing, route count, context length, measured performance when available, and links to each model's provider and benchmark pages.