Your Data. Your Control. Our Promise.

Our Golden Rule: We Do Not Train on Your Data.

We never use the content of your Inputs or Outputs to train any AI models—not our own, and not those of third parties. This promise is the core of our service and is legally baked into our Terms of Service and Data Processing Agreement.

Data Ownership

You keep full ownership of your Inputs. We, in turn, assign all our rights to the Outputs generated for you by the Service. Simply put: what you create is yours.

Data Retention: Radical Transparency

• Our Logs: For debugging, billing, and security, we temporarily log operational metadata and the content of your Inputs and Outputs for a maximum of 30 days. After this period, they are automatically and permanently deleted from our systems.
• Our Partners’ Logs: As a proxy service, your data is also processed by our AI model providers. Most of our partners, like OpenAI and Anthropic, also have a default 30-day retention policy for monitoring abuse. We are transparent about this so you can make an informed decision.
• Zero Data Retention (ZDR): For customers with the highest privacy needs, our Enterprise plan offers a Zero Data Retention option, where available from our partners (like Mistral AI ). This ensures your data is not stored after processing.  

Data Residency

Our primary servers are in Germany, within the European Union. When we process your requests, we prioritise using EU-based endpoints offered by our partners, such as those from OpenAI and Anthropic. However, depending on the provider, some processing may occur in the United States. For a full breakdown, please see our list of sub-processors.