Explain complex ideas
in one smooth take.
A native macOS infinite canvas for tech creators, educators, and speakers. Lay out cards and code, let AI draft boards via built-in MCP, and glide the camera step by step as you record or export offline HTML.
Rust + GPUI · Built-in MCP & Skill · 1:1 Offline HTML Export · Windows waitlist open
Built with Telluma · This interactive canvas is a real 89 KB standalone HTML export (⇧⌘E). Click or press ← → to step, hold F for overview.
See how HTML Export works ↓Hand your outline to an AI agent. Open a finished board.
Don't drag boxes by hand for an hour — and don't let LLMs guess broken JSON coordinates. Telluma ships a native stdio MCP server (Telluma --mcp) and official Agent Skill powered by the app's real layout and CoreText measurement engine.
- 1Zero-overlap auto-layout engine
Horizontal flow, vertical reading flow, or 360° radial spread — measured with macOS CoreText so every card breathes cleanly.
- 2Bundled Agent Skill with live template discovery
Teaches your agent narrative pacing and card restraint, while querying
list_templatesdynamically so it never goes stale. - 31-click setup inside Telluma Settings
One button wires both MCP and Skill for Antigravity, Codex, Claude Code, or Cursor — global or per-project.
One self-contained HTML file. Present anywhere, offline.
The interactive guide you just clicked through at the top of this page? That's a real ⇧⌘E export — images, syntax highlighting, and smooth camera physics packed into one file with zero external dependencies.
Live talks & classrooms: Open in any browser without installing Telluma — wireless presenter clickers work right away.
Share directly with viewers: Send one file in chat or email; phones and tablets get native pinch-to-zoom and swipe gestures.
Embed on your site or blog: Drop interactive visual explainers straight into articles and docs — just like this landing page.
Fast, quiet, and built specifically for recording.
One-take screen recording
Snap to max 16:9 (⌥⌘R), auto-hide titlebar and cursor as you step, keep your channel signature pinned in the corner, and hold F to peek overview anytime.
Frameless Markdown & code highlight
Double-click any card for live frameless Markdown, Tree-sitter syntax highlighting, smart ⌘V paste for images & code, and single-file .telluma archives.
120 FPS Retina, paper-like calm
Native CoreText rendering in a ~17 MB macOS app — zero Electron bloat, paired with 9 muted paper card tones.
Join the Early Access Waitlist
Telluma is currently in early access. Leave your email below and I'll reach out personally as new spots open — starting with macOS (Apple Silicon).
Questions & Answers
What stage is Telluma in right now?
Telluma is in early access. I am onboarding creators in small batches so I can gather real workflow feedback directly and refine the experience before a public release.
Which operating systems are supported?
Early access starts with native macOS (Apple Silicon). The core engine and .telluma file format are built to be cross-platform, so Windows users are welcome to join the waitlist to hear first when it lands.
How is Telluma different from Keynote, Miro, or Excalidraw?
Slide decks chop a story into disconnected frames; whiteboards are great for brainstorming but awkward to record cleanly. Telluma pairs spatial layout with a dedicated Play Mode — step-by-step reveal, smooth logarithmic camera flight, a 16:9 recording frame, and zero UI clutter.
How does the AI Agent / MCP integration work?
Telluma ships with a local stdio MCP server (Telluma --mcp) and a ready-to-install Agent Skill right inside Settings. Connect Antigravity, Codex, Claude Code, or Cursor and let your agent lay out a zero-overlap board directly from your notes or codebase.
What happens after I join the waitlist?
I read every signup personally and send invites out in small batches from support@telluma.app. Sharing what kind of videos, articles, or talks you make helps me prioritize early testers.