Inference API
OpenAI-compatible API for open-weight models. Existing SDKs and tools keep working; you only change the base URL and the API key.
AI and LLM hosting in Germany
Use language models without sending your data to US AI providers. We host open-weight models on our infrastructure in Germany and make them available through an OpenAI-compatible API and a chat interface for your team.
Alpha offering
This offering is at an early stage. Models, interfaces and conditions may still change. Ideal for evaluation and pilot projects.
What you get
Open-weight models
Currently models from the Qwen family for chat, coding and reasoning. Further open-weight models on request.
AI chat for your team
A chat interface in the browser based on Open WebUI, connected to the hosted models. On request with single sign-on via Managed Keycloak.
Dedicated model instances
Your own model deployment that is not shared with other customers. Scope and sizing on request.
Who it is for
- Companies that want to use generative AI with data that must not leave the EU
- Development teams that build AI features into their products
- Teams that want an internal AI chat instead of public AI services
Frequently asked questions
What does alpha mean?
The offering is at an early stage. Models, interfaces and conditions may still change. It is well suited for evaluation and pilot projects; describe your use case in the request.
Is my data used to train models?
No. We run inference only and do not train models with your inputs. Prompts and answers are processed on our infrastructure in Germany.
Which models are available?
Currently models from the Qwen family. Tell us which open-weight model you need and we check whether we can host it for you.
Can I use it with my existing tools?
Yes. The API is OpenAI-compatible, so libraries and tools that support the OpenAI API can be connected by changing the base URL.
AI agents
AI employees with staffplane
Beyond models, we also host AI agents: digital team members that take on recurring tasks in your tools. We offer this under our brand staffplane, together with Staffy, its friendly copilot.
Combine with
Request a quote for AI & LLM hosting
Describe your requirements. You receive an offer tailored to your project.
Request a quote