Built for teams
Cursors, edits, and generations sync live for everyone in the workspace
Feature
Open canvasCanvas
One infinite surface for the whole process. Drop a node, pick any image, video, or audio model, and wire outputs into the next prompt until the work lands exactly as you imagined.
Features
Tools built in
Edit, inpaint, upscale, crop, and resize without leaving the canvas
Every node type
Text, image, video, and audio, each running the model you choose
Ask Mel
Describe what you want. Mel picks the models and builds the nodes.
Frequently asked questions
What is a node-based canvas?
Every step of the work is a node with its own prompt, model, settings, and version history. Edges pass the output of one node into the input of the next, so a single generation becomes a pipeline you can rewire, branch, and reuse.
Which models can I use on the canvas?
Image, video, audio, and text models from every major provider, under one subscription — GPT Image, Nano Banana, Flux, Seedream, Ideogram, Veo, Kling, Seedance, ElevenLabs, and more. You pick the model per node, or let Mel choose.
Can my team work in the same canvas at once?
Every canvas is multiplayer by default. Cursors, edits, runs, comments, and sticky notes sync live, and AI agents appear as participants with a highlighted working area.
