Index / providers / nebius

Nebius Token Factory

Pay-per-token serverless inference from European cloud company Nebius, serving Nous Research's Hermes 4 405B among mainstream open-weight models.

Models1

Per-token price$1.50/1M tok

Plans from—

Free tierNo

tokenfactory.nebius.com ↗Pending review

[ 01 ] Models & prices

BILLINGPay as you go per token. A one-time $1 trial credit (card required). Companies can pay by bank transfer.

Model	Method	Context	In $/1M	Out $/1M	Per 1K chats
NousResearch/Hermes-4-405BLlama 3.1 405B · 405B · Llama 3.1 Community	Fine-tuned	131K	$1.00	$3.00	$2.50

Per 1K chats = cost of 1,000 messages of ~1,000 tokens in and ~500 out. Click a model to compare it across providers.All models →

[ 02 ] At a glance

Method: Fine-tuned
Access: Web chat + OpenAI-compatible API
Logging: Retained
Trains on prompts: No
Guarantee: Policy only
Jurisdiction: Netherlands, EU
Signup: Account required
Payments: Card, Other
Data checked: Oct 11, 2026

PRIVACYBy default prompts and outputs are retained for speculative decoding (not training); organizations can opt into zero data retention. Retained data is stored in Finland. Policy ↗

[ 03 ] Scores

Compliance

Benchmark pending

measured

Capability

Benchmark pending

measured

Privacy

38.0

from published policy & pricing

Truthfulness

Benchmark pending

measured

Cost

43.2

from published policy & pricing

Speed

Benchmark pending

measured

Composite: pending capability and compliance runs · method

[ 04 ] Notes & sources

Formerly Nebius AI Studio. Hermes 4 70B moved to dedicated endpoints only on Aug 31, 2026. OpenRouter routes Hermes 4 405B here.