The Known Good Updated 25 Jul 2026
Providers· Inference host· 13 tracked models

Nebius — pricing & performance

Compare models All providers Data sources
Models served
13
of 908 tracked models
Mean output speed
63
tokens per second, averaged over
13 measured endpoints
Mean latency
0.5s
time to first token, averaged over
13 measured endpoints
Cheapest blended
$0.10
per 1M tokens at a 3:1
input:output blend
Output speed on Nebius

Median output tokens per second for each model this host serves, derived from OpenRouter endpoint telemetry. Higher is better. We do not run these measurements ourselves.

12 of 13 models ▾ ☰⚙⇩ +Add model from specific provider
281
115
99
84
52
34
33
28
28
26
25
12
The Known Good
13 models on Nebius
Full leaderboard →
Model Creator Input $/M Output $/M Blended Speed TTFT Context
Google: Gemma 3 27B Google $0.10 $0.30 $0.15 12 0.7s 110k
Meta: Llama 3.3 70B Instruct Meta $0.13 $0.40 $0.20 34 0.3s 131k
Nous: Hermes 4 405B Nous Research $1.00 $3.00 $1.50 33 0.2s 131k
Nous: Hermes 4 70B Nous Research $0.13 $0.40 $0.20 84 0.2s 131k
NVIDIA: Nemotron 3 Nano 30B A3B NVIDIA $0.060 $0.24 $0.10 28 1.2s 262k
NVIDIA: Nemotron 3 Super NVIDIA $0.30 $0.90 $0.45 3 0.7s 262k
OpenAI: gpt-oss-120b OpenAI $0.15 $0.60 $0.26 281 0.2s 131k
Qwen: Qwen2.5 VL 72B Instruct Alibaba $0.25 $0.75 $0.38 25 0.8s 32k
Qwen: Qwen3 235B A22B Instruct 2507 Alibaba $0.20 $0.60 $0.30 52 0.4s 262k
Qwen: Qwen3 30B A3B Instruct 2507 Alibaba $0.10 $0.30 $0.15 115 0.3s 262k
Qwen: Qwen3 32B Alibaba $0.10 $0.30 $0.15 28 0.5s 41k
Qwen: Qwen3 Next 80B A3B Thinking Alibaba $0.15 $1.20 $0.41 99 0.3s 128k
Z.ai: GLM 5.1 Z.ai $1.40 $4.40 $2.15 26 0.5s 203k
The Known Good
Source. Prices, endpoint availability and throughput for Nebius are ingested from OpenRouter; model metadata is enriched from models.dev. A blank cell means the figure was not published for that endpoint — it never means zero. Blended price is (3 × input + output) ÷ 4, the same 3:1 blend used everywhere on this site; see methodology. Quality scores are not provider-specific: a model scores the same wherever it is hosted, so quality lives on the model pages.