Design ads, posters, and flyers at the speed of AI

A layered image canvas — text, shapes, pen, shadows, masks and over 80 fonts — with an AI assistant that sequences hundreds of edits in a single run.

Canvas — Untitled Design
Elm & VineOct
WINENIGHTFri 18 Oct · 7–10 PMElm & Vine Bar · 21+
☰ASSISTANT+

Background and ornament imagery is generating on the canvas now.

I filled in placeholder details so the first draft is complete — tell me the real date, venue and age limit and I’ll drop them in.

ChatEdit
Describe the change you want…↑

What you can design

Posters and print-style ads

Build a layout as real layers — ground, cutout, headline, credits — that all stay editable in the image canvas

Cutouts and background removal

Lift a product off its background in one click and drop it onto a ground made with AI image generation

Type-led layouts

Set headlines in over 80 display and text families, with alignment, shadows and opacity, in the design editor

Mask images into type

Cut a photograph to the shape of a headline or a vector path, and keep editing both, in the AI image editor

Take a still into motion

Move a finished layout onto a multi-track timeline and export MP4 from the video editor

Finish remixed ads

Fix the type and composition of ads built from live campaigns with Ad Remix

Frequently asked questions

How does the AI assistant work?+

You describe what you want in plain text and the assistant builds it on the canvas as real layers — type, shapes, masks and generated imagery — rather than handing back a flattened image. It plans the whole composition before it starts, drawing on a large reference library of print and editorial design, then carries that plan out as a sequence of edits you watch land on the page. Everything it makes stays editable, so you can move a headline or change a colour afterwards without regenerating anything.

How is this different from Figma or Canva?+

The canvas will feel familiar — layers, type, shapes, masks, a details panel. The difference is how much the assistant does per instruction. Most design tools’ AI performs one step and hands control straight back to you. Ours plans an entire composition and executes it unsupervised: often more than a hundred operations across type, shapes, masks and generated imagery in a single run, over minutes rather than seconds. You watch it work and step in when you want to, instead of approving every step.

Is the AI image editor free?+

Yes. The canvas is free to use in the browser, with every editing tool and every font available. Paid plans add monthly credits for AI image generation and commercial usage rights.

What editing tools does the canvas have?+

Text with full typographic control, shapes, a pen tool for vector paths, drop shadows, opacity, and masking — cut any image to the shape of type or a vector. Layers can be reordered, hidden and restyled, and background removal and image generation run in the same document, so you never export to do one step somewhere else.

What fonts can I use?+

Over 80 families are installed, from editorial serifs and grotesques to display weights and hand-lettered faces. The assistant chooses a typeface for each block of type as part of the layout, and you can override it on any layer.

Do I need design experience?+

No. Most people describe what they want and let the assistant compose it — because it works from real print references rather than a blank page, the result is a considered layout instead of a template with your words dropped in. The full editing tools are there for when you want to take over.