Nebius AI Studio & Cloud Setup: Token API & InfiniBand Guide (2026)
IntermediateEstimated setup time: 4 minutes•Updated for 2026 infrastructure
KYC Policy
Credit Card Authorization & Business Verification
Payment Methods
Credit / Debit Card • Enterprise Invoicing
Min. Deposit
$0 (With promo trial credits available)
Billing Model
Per-second billing for VMs; Per-token for AI Studio API
Trial Credits
Free trial grants for verified developer accounts
Terminal Access
Dedicated SSH, Managed Slurm, and Kubernetes (k8s)
What You Need Before Starting
✓Corporate or GitHub email
✓A valid credit card for account tier activation
Step-by-Step Setup Walkthrough
3 Steps to First Run1
Sign Up for Nebius AI Cloud & Studio
Step 1 of 3Create your organization and project in the European sovereign cloud portal.
- •Sign up at nebius.com with your professional email.
- •Set up your Organization and default billing account.
2
Accessing Serverless Inference vs GPU Virtual Machines
Step 2 of 3Nebius offers both 1-click token APIs (AI Studio) and full H100/H200 InfiniBand clusters.
- •For instant LLM inference: Generate an API key in 'AI Studio' and use the OpenAI-compatible endpoint.
- •For full compute clusters: Go to 'Compute' → 'Create VM' with 3.2Tbps InfiniBand networking.
Nebius AI Studio OpenAI Drop-in Code
python
from openai import OpenAI
client = OpenAI(
base_url="https://api.studio.nebius.ai/v1/",
api_key="YOUR_NEBIUS_API_KEY"
)
response = client.chat.completions.create(
model="meta-llama/Llama-3.3-70B-Instruct",
messages=[{"role": "user", "content": "Hello Nebius!"}]
)
print(response.choices[0].message.content)3
Configure VPC & SSH for InfiniBand Clusters
Step 3 of 3Set up security groups and direct SSH access to dedicated H100 clusters.
- •Select your H100 SXM5 or H200 SXM nodes with InfiniBand fabric.
- •Upload your SSH key and attach a public IP address.
- •Connect directly to verify GPU status using `nvidia-smi`.
Insider Cost-Saving & Anti-Waste Tactics
#1: Use Nebius AI Studio token billing for variable inference rather than spinning up full idle H100 VMs.
Frequently Asked Questions (Nebius AI Studio Onboarding)
Q:Is Nebius AI Studio compliant with European GDPR?
Yes. Nebius datacenters are located in Europe (e.g. Finland, France) and fully adhere to European data sovereignty and GDPR compliance rules.
Q:Can I use Nebius as a direct drop-in replacement for OpenAI API?
Yes. Nebius AI Studio provides an OpenAI-compatible endpoint for Llama 3.3 70B, DeepSeek-R1, and Qwen models with ultra-low latency.
Official Developer Onboarding Channel
Ready to Launch on Nebius AI Studio?
Followed the guide? Jump directly to the official console using our verified partner route with guaranteed zero markup and active referral tracking.