Canva’s first AI image model plans designs before generating them

Canva has launched its first AI image generation model, called Canva Craft, which competes with models from OpenAI, Microsoft, and Google. Craft uses a two-stage architecture that lets it plan a design before generating the image, providing greater precision, consistency, and control. It can generate images at up to 2K resolution and in more than 30 aspect ratios, for uses such as social media posts, marketing content, realistic scenes, and digital art. The two-stage system helps it accurately render complex images that require legible text and precise object placement. The model builds on Canva’s broader AI push, including prompt-based editing and layered AI-generated images launched in April. Canva says Craft was trained on millions of human-made templates from its content library, such as social media posts, presentations, infographics, and menus. It currently ranks fourth on Design Arena’s leaderboard for AI image generation models, behind OpenAI’s GPT-Image-2.5 Sunburst, GPT-Image-2.5 Flare, and GPT-Image-2. Canva says Craft includes robust input and output moderation, privacy and safety controls, and reporting tools, and plans to support AI agents through the open Model Context Protocol as well as future Craft-fast and Craft-ultrafast models. Canva Craft is available now globally, accessible through the three-dot option on Canva’s left-hand menu and then Dream Lab.

(Source：The Verge)