Your data stays in the EUProcessed in European data centres, never used for training, covered by one DPA.Talk to sales
Popular models
All models (35)
Change one linePoint the OpenAI SDK at our base URL and keep the rest of your code.
```
client = OpenAI(
  base_url="https://api.inference.eu/v1",
  api_key=INFERENCE_EU_API_KEY,
)
```
Coding agents and IDEs
Aider
Claude Code
Cline
Continue
Cursor
Grok CLI (Grok Build)
SDKs and frameworks
Anthropic SDK (Python and TypeScript)
CrewAI
Haystack
LangChain
LiteLLM
LlamaIndex
Chat apps and automation
AnythingLLM
Dify
Flowise
goose
Hermes Agent
LibreChat
All integrations (32)
Claude Code on EU-hosted modelsSet two environment variables and Claude Code runs on open models processed in the EU.Read the setup guide
API endpoints
Chat completions
Responses
Messages (Anthropic format)
Count tokens
List models
Embeddings
Rerank
Speech to text
Guides
Migrate from OpenAI
Streaming
Tool calling
Structured outputs
Vision
Reasoning models
Prompt caching
Rate limits and errors
Tools
Document conversion
Remote MCP servers
Web search
Code execution
API overview
Your first requestSend a chat completion with curl and your API key.
```
curl https://api.inference.eu/v1/chat/completions \
  -H "Authorization: Bearer $INFERENCE_EU_API_KEY" \
  -d '{"model": "gpt-oss-120b",
       "messages": [{"role": "user",
                     "content": "Hallo"}]}'
```

EN

Get access to EU-hosted AI models and agents

Tell us what you're building. We work with companies, so please use your work email.

1. Tell us what you're building
Which models you need, roughly how much traffic, and any compliance requirements.
2. We reply within one working day
With answers to your questions, pricing for your volume, and a call if it helps.
3. You get an API key and a DPA
Start calling models straight away, with a data processing agreement ready to sign.