Hetzner Inference API

✅ Truly Free

Hetzner (a major EU cloud/hosting company) runs an experimental, OpenAI-compatible inference API launched in July 2026. It is free of charge while experimental and serves Qwen3.6 35B A3B and Qwen3.8 27B (262K context, text and image) on EU servers. You sign in with a regular Hetzner account, which needs full billing details; several users report that a credit card was requested at account creation (Hetzner may ask for a card or ID to verify new accounts), although inference itself is not billed. Limits: 10 requests and 4M input / 100K output tokens per minute per key. No SLA.

Truly Free Experimental EU Hosted No Card
Get API Key Suggest Edit
1

Overview

Provider Type

API

API Endpoint

https://inference.hetzner.com/api/v1

Free Tier Highlights

3M input / 60K output tokens per 60s

Why Choose Hetzner Inference API?

Hetzner Inference API stands out for its unique features and capabilities. With a developer-friendly API and comprehensive documentation, you can integrate AI capabilities into your applications within minutes.

Quick Start Guide

1

Create a Hetzner account

2

Visit https://experiments.hetzner.com/inference

3

Generate an API token

4

Configure client with base_url https://inference.hetzner.com/api/v1

Available Models

Model Name ID Context Capabilities
Qwen3.8 27B Free
Qwen3.8-27B
262 000 Context
-
Qwen3.6 35B A3B Free
Qwen/Qwen3.6-35B-A3B-FP8
262 000
Vision Reasoning

Integration Examples

Ready-to-use code snippets for your applications.

main.py
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_HETZNER_API_TOKEN",
    base_url="https://inference.hetzner.com/api/v1"
)

response = client.chat.completions.create(
    model="Qwen3.8-27B",
    messages=[
        {"role": "user", "content": "Hello!"}
    ]
)

print(response.choices[0].message.content)

Free Tier Pricing & Limits

Rate Limit

Requests per minute

10 requests4 000 000 input / 100 000 output tokens per 60s (per API key)

Daily Quota

Requests per day

No daily cap published (free while the API is experimental); a Hetzner account with billing details is required and Hetzner may ask for a card or ID to verify it

Token Limit

Tokens per minute

See rate limit

Monthly Quota

Per month limit

Free during experimental phaseno billing system yet

Use Cases

EU data compliance testing

High-volume experimentation while it lasts

Secondary/backup free API

Limitations & Considerations

Experimental — no SLA, could be discontinued or paywalled at any time

Only one model currently offered

No official uptime/support guarantees

Community Hub

Live

Join the discussion, share tips, and rate Hetzner Inference API.

Quick Reactions

Add Discussion

Comments are moderated. Be helpful and respectful.

Recent Activity

0 comments

Ready to Get Started?

Join thousands of developers using Hetzner Inference API

Start Building Now