Canva has announced its first proprietary artificial intelligence image model, entering direct competition with major technology companies such as OpenAI, Microsoft, and Google. According to reporting by The Verge, the new model is named Canva Craft and features a two-stage architecture designed to plan layouts prior to rendering pixels.
How the Two-Stage Architecture Works
Standard text-to-image generators typically attempt to create an entire visual composition in a single pass based on a user prompt. Canva Craft differs by breaking the process into distinct phases. The system plans the design structure before generating the final image output. The company states this architectural approach delivers greater precision, consistency, and control compared to traditional single-pass generation methods.
Unknown Details and Market Position
Despite the announcement, significant details regarding the model remain undisclosed. The public rollout timeline, specific technical parameters, and pricing structure were not specified in the release. Observers will be watching to see how this tool integrates into Canva’s existing suite of design software and whether it will be offered to all user tiers.
As the generative AI space becomes increasingly crowded with enterprise offerings, Canva’s entry signals a shift toward specialized tools tailored for layout workflows.
Lead image: https://kaboompics.com/ / Pexels
