Language models
Chat, drafting and reasoning from models we host ourselves. It's the engine behind Aurora, our AI assistant for teams.
Moonly Gateway is the private front door to our self-hosted AI models. One gateway for language, embeddings and reranking. Run entirely on European infrastructure, with no data sent to third-party model providers.
Our products shouldn't each wire up their own models. Gateway gives them a single, controlled way to use AI, and gives us one place to keep it safe.
Chat, drafting and reasoning from models we host ourselves. It's the engine behind Aurora, our AI assistant for teams.
Turn your data into vectors so apps can search by meaning, not just keywords. Your documents are embedded without leaving our servers.
Sharpen search results by putting the most relevant answer first. It's the difference between "found something" and "found the right thing".
Every request is handled on infrastructure we operate in the Netherlands. There's no hidden hop to a US cloud and no prompt used to train someone else's model.
Gateway runs Aurora's assistant, and gives our other products semantic search and reranking over their own data.
Building on Moonly? Ask us about connecting your product to Gateway.
Gateway access is provided to Moonly products and selected partners. Tell us what you're building.
Get in touch