ARTIFICIAL INTELLIGENCE · FOCUS
Canva launches Craft, a model that plans image composition
Canva announces an image generation model built around a design planning phase. Craft is available worldwide, according to The Verge. Improvements in text accuracy and object placement remain performance claims made by the company.
Legal disclosure: this content was produced by an artificial intelligence system. Editorial transparency
Canva is launching Craft, its first image generation model, with an architecture that plans the design before producing the image. According to The Verge, the announcement was made on Thursday. Canva credits this structure with greater precision, more consistency and increased control over composition.
The model can generate images at resolutions up to 2K and in more than 30 aspect ratios, according to the reported specifications. Announced uses range from social media posts to marketing content, realistic scenes and digital art. Canva highlights text readability and precise object placement in complex compositions in particular. These benefits remain company claims: the available information does not constitute an independent assessment of their reliability.
Canva says it trained Craft on millions of human-created templates from its content library. These include presentations, infographics, social media posts and menus. This stated origin describes the training material; it is not sufficient to establish the quality of every image produced.
According to The Verge, Craft is available to all users worldwide through Dream Lab. The practical value of the approach will depend on its ability to meet compositional constraints. For the uses mentioned, better text and object placement could reduce the corrections needed, but that benefit remains to be verified.