What is Lunar?

Lunar is an AI learning companion. It looks like a chat, but it is built around a pedagogical idea: a good answer is only the first step of learning. What you cover in a conversation should become material you can revisit, validate and master — and it should still be there next week.

Two layers, one conversation

A regular chatbot replies and forgets. Lunar works on two layers at once:

• The conversation layer — you ask, Lunar explains, and the thread keeps its full context so follow-ups build on what was already said.
• The study layer — while you talk, Lunar organizes what you cover into interactive citations, microchats, a concept tree, study anchors and a study map.

The second layer is what makes Lunar a learning tool rather than a chat window: ideas stop being messages that scroll away and become material you can review and test.

What makes it different

Capability	What it gives you
Interactive citations	Highlighted terms in every answer open a side panel with a contextual definition and follow-up questions — no tab-switching to look things up.
Concept tree	A living map of the ideas you have covered and how they relate, per chat and across your whole Space.
Spaces	Long-lived containers for a degree, subject or exam, with their own memory, files and progress.

Built to follow your semester

Lunar organizes your work around long-lived Spaces instead of one-off chats. Your classes, files, accumulated memory and progress live in the Space, so what you studied last week is still available today — and a new chat starts already knowing what came before it.

[!TIP] New to Lunar? Read Quick start next — it walks you from sign-up to your first study session in a few minutes.

About this documentation

These pages explain how Lunar behaves in real use: how to start, which model to choose, how citations, the concept tree, microchats and Spaces work, and what to do when something does not behave as expected. Use the sidebar to browse, or search to jump straight to a topic.