Canva releases first image generation model called Canva Craft

Model and generation process

Canva has released its first image generation model, called Canva Craft. The model was developed by the company's research team to compete with image generation models from Google, OpenAI, ByteDance, and Meta. According to Canva, the model works in two stages. In the first stage, it interprets a user's prompt and creates a plan that includes spatial relationships of elements, relative scale, visual hierarchy, and information for text generation. In the second stage, it generates an image of up to 2K resolution in one of 30 possible aspect ratios for various platforms. Canva said this approach can produce images with accurate details of objects and legible text, and it can allow precise edits without changing other parts of the image.

Training and design features

Canva said it used designs rather than photographs to train the model. The training used human-designed templates from Canva's library, including posters, social posts, presentations, infographics, menus, and typography-rich layouts. Because of this, images generated with Canva Craft can be converted into layered designs using Canva Magic, the company's image-to-design transformer. Canva also said its platform released a model last year that understands designs and layers.

Availability and business context

Canva Craft is available on the company's platform to all users and can be accessed from the Dream Lab tab. The model will also be available through the Canva API. Canva plans to support image generation in third-party chatbots such as ChatGPT and Gemini through Canva's MCP. The company has used several image models for generation on its platform, so the new model may help save API costs. Canva said the new model also allows better quality and speed control. Researchers from Leonardo.ai, which Canva acquired in 2024, contributed to the new image model. Canva is also developing two cost-efficient versions, called Craftfast and Craft-ultrafast, which will be released soon. The company was on track for $4 billion in revenue last year, but it cut its growth forecast for 2026. External analysts reportedly said its valuation fell from $42 billion to $35 billion.

Sources: TechCrunch