Canva Releases Craft Image Generation Model

Canva has released Canva Craft, its first image generation model, according to TechCrunch. The model is available to all Canva users through Dream Lab and will also be offered through the Canva API.

Canva Craft first interprets a prompt and plans the spatial relationships, scale, visual hierarchy, and text. It then generates an image at up to 2K resolution in one of 30 aspect ratios. Users can edit specific elements without changing other parts of the image.

Canva trained the model on human designed templates from its library, including posters, presentations, infographics, menus, and layouts with extensive typography. Generated images can be converted into layered designs using Canva Magic.

Researchers from Leonardo.ai, which Canva acquired in 2024, contributed to the model. Canva also plans to support generation through third party chatbots such as ChatGPT and Gemini using its MCP, and is preparing faster versions called Craftfast and Craft-ultrafast.