Cluster Online | - nodes · - GPUs · -GB VRAM

Private AI API
OpenAI-Compatible

Drop-in replacement for OpenAI. 18 models, GPU-accelerated, hosted in the Netherlands. Your data stays private. GDPR compliant.

from openai import OpenAI

client = OpenAI(
    base_url="https://api.helheim-ai.dev/v1",
    api_key="YOUR_API_KEY"
)

response = client.chat.completions.create(
    model="qwen",
    messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)
🔄

Drop-in Replacement

Same API as OpenAI. Change one line of code — your base_url — and you're running on private hardware.

GPU Accelerated

Multi-GPU CUDA inference with 28GB VRAM. RTX 3060 + RTX 5060 Ti. Models load from RAM in <1 second.

🔒

Private & GDPR

Hosted in the Netherlands. Your data never leaves our network. No logging, no tracking, no third-party cloud.

Available Models

18 models ready for inference. Smart routing picks the best one for your task.

Qwen 2.5 14B Best Allround
model="qwen"
14B params
Qwen 2.5 Coder 14B Best Code
model="code"
14B params
Mistral Small 24B Largest
model="mistral-small"
24B params
DeepSeek R1 14B Reasoning
model="deepseek-r1"
14B params
Phi-4 Microsoft
model="phi"
14B params
Gemma 2 9B Google
model="gemma"
9B params
Mistral 7B v0.3 Fast
model="mistral"
7B params
StarCoder2 15B Code
model="starcoder"
15B params

+ 10 more models. Use model="auto" for smart selection or model="fast" for lowest latency.

Quick Start

Three steps to your first API call.

1

Get an API Key

Sign up via the dashboard or contact us for an API key.

2

Install the SDK

pip install openai
Works with the standard OpenAI library.

3

Make a Request

Set base_url to api.helheim-ai.dev/v1 and you're live.

API Endpoints

POST /v1/chat/completions Chat completion (OpenAI format)
GET /v1/models List available models
GET /health Cluster health check

Simple Pricing

No hidden fees. Pay for what you use.

Free Tier

Try it out

€0
  • 100 requests/day
  • All models
  • OpenAI-compatible
  • Priority queue
Get Started
Popular

Developer

For indie devs & startups

€19/mo

~10,000 requests included

  • Unlimited requests
  • All models
  • Priority queue
  • Usage dashboard

Business

For teams & companies

€79/mo

~50,000 requests included

  • Everything in Developer
  • Dedicated GPU time
  • Custom models
  • Priority support

Why Helheim?

Helheim AI OpenAI Groq
Privacy / GDPR ✓ EU hosted ✕ US ✕ US
Data logging ✓ None ~ 30 days ~ Unknown
OpenAI-compatible
Starting price Free $20/mo Free (limited)