Canva releases its first in-house image model, Canva Craft
Canva has launched its first in-house frontier image model, Canva Craft, which ranks second on the DesignArena image generation leaderboard, behind only OpenAI. The model consists of two parts: a planner and a renderer. The planner first writes out a composition plan, and then the renderer draws it at up to 2K resolution and in more than 30 aspect ratios. Because the layout and text are determined before rendering, when a single element is modified, only that element changes.
Why read this
AI score 74
The original article breaks down the design that divides responsibilities between the Canva Craft planner and renderer, so readers can understand how layouts become explicit intermediate objects.
Topics
View original postIntroducing Canva Craft 🚀 Extraordinary image generation is now available to everyone, powered by our first frontier model built specifically for design. With Craft, Canva is now the world's second most advanced AI lab for image generation on @DesignArena's global leaderboard. See what it can do 👇
View original
Introducing Canva Craft 🚀 Extraordinary image generation is now available to everyone, powered by our first frontier model built specifically for design. With Craft, Canva is now the world's second most advanced AI lab for image generation on @DesignArena's global leaderboard. See what it can do 👇
Body
Canva has entered the frontier AI model race, and its first model is already near the top.
It's first in-house image model made it the second-best (on DesignArena) lab in image generation, with only OpenAI ahead of it.
Canva Craft has a planner that reads the prompt and writes out the composition before any pixels exist: which elements appear, where each one sits, how large it is, what carries the emphasis, and the exact words to display.
A second model, the renderer, paints that plan at up to 2K across more than 30 aspect ratios.
So layout is no more something that the model has to get right inside a single pass. It becomes an explicit intermediate object.
Spelling gets settled before rendering starts. The renderer’s job is to draw words it was already given, in a spot it was already assigned.
The plan pays off a second time during revision. Change one element and only that element changes, so fixing a subhead leaves the product shot next to it alone.