Canva releases Craft, its own image model that plans a layout before drawing
Canva launched Canva Craft, the first image model built and trained from scratch by its in-house lab, Canva Research. It works in two steps: one model writes a plan of the image (what goes where, how big, and any text), then a second model renders it at up to 2K resolution in more than 30 aspect ratios. Canva says this gives more legible text and lets you change one element without redoing the rest, and that Craft debuted fourth among image models on DesignArena, a crowdsourced design ranking. It is available now to all Canva users through Dream Lab and to developers through Canva's public API, with Canva MCP access for AI assistants listed as coming soon.
Why it matters: If your team makes posters, social posts or slides with text on them, try Craft in Dream Lab on a few real briefs and check the text spelling before you use it. The ranking is Canva's own claim from a crowdsourced leaderboard, not an independent test.
Canva ↗AI-assisted, reviewed by Sorted AI
▲ 0
0 comments