What changed
Design platform Canva released its first image generation model, Canva Craft, developed by its research team to compete with same-category models from Google, OpenAI, ByteDance, and Meta. The model works in two stages: first interpreting user prompts to create a plan with details like spatial relationships, relative scale, and visual hierarchy, and then generating an image of up to 2K resolution in 30 different aspect ratios.
What this means in practice
Users can now generate images with accurate object details and legible text, and make precise edits without altering other parts of the image. The model is trained using human-designed templates from Canva’s library, including posters, social posts, and infographics. Generated images can be converted into layered designs using Canva Magic. Canva Craft is available on the platform via the Dream Lab tab and through the Canva API. The company also plans to support image generation in third-party chatbots like ChatGPT and Gemini through its MCP. Canva has used multiple image models on its platform, and the new model is expected to reduce API costs and improve quality and speed control.
Limitations and availability
Canva Craft is available to all users on the platform and through the Canva API. The company is also developing two cost-efficient versions, Craftfast and Craft-ultrafast, which will be released soon. The model uses designs rather than photographs for training, which may affect the types of images it can generate. The supplied announcement does not specify the exact number of parameters or resolution details beyond 2K.
