Canva has introduced its 1st in-house AI image generation model, Canva Craft, making it available to users worldwide. The model is designed to create high-quality visuals with clearer text, accurate object placement and greater control over image layouts.
Canva Craft uses a 2-stage image generation process
Unlike conventional image generators that create images directly from prompts, Craft follows a 2-stage approach. It first interprets the prompt and plans the composition, including the placement, size and visual importance of elements, as well as any text. A separate rendering model then converts this plan into an image.
The model can generate images at up to 2K resolution and supports more than 30 aspect ratios. It can also generate text in 55 languages, making it suitable for social media posts, posters, presentations, infographics, marketing materials, portraits, landscapes and digital art.
Canva says this architecture improves text readability and allows users to make targeted changes without unnecessarily altering other parts of an image. The generated visuals can also work with Canva’s Magic Layers feature, which converts images into editable design elements.
Trained on human-created designs
Craft was trained using millions of human-made templates from Canva’s content library, including posters, social media graphics, presentations, menus and typography-focused layouts. The model was further refined using supervised fine-tuning and reinforcement learning, with feedback from professional designers.
For internal testing, Canva’s designers and researchers assessed more than 25,000 pairs of images against competing models. The company says designer feedback took priority when it differed from automated evaluations.
Craft ranked 4th overall on DesignArena’s image-generation leaderboard at launch. Its results ranked 1st in People & Portraits and Landscapes & Nature, 2nd in Abstract & Patterns, and 4th in Products.
Global availability and future plans
Canva Craft is available to all users worldwide through the Dream Lab section of the platform. Developers can also access it through Canva’s public API to integrate image generation into other products and services.
Canva is developing 2 additional versions, Craft-fast and Craft-ultrafast, designed to deliver high-quality images at lower costs. Support for AI assistants and automated workflows through Canva’s MCP Server is also planned.
The model is backed by Canva Shield, which includes content moderation, privacy and safety controls, ongoing testing and reporting tools.
Also read: Viksit Workforce for a Viksit Bharat
Do Follow: The Mainstream LinkedIn | The Mainstream Facebook | The Mainstream Youtube | The Mainstream Twitter
About us:
The Mainstream is a premier platform delivering the latest updates and informed perspectives across the technology business and cyber landscape. Built on research-driven, thought leadership and original intellectual property, The Mainstream also curates summits & conferences that convene decision makers to explore how technology reshapes industries and leadership. With a growing presence in India and globally across the Middle East, Africa, ASEAN, the USA, the UK and Australia, The Mainstream carries a vision to bring the latest happenings and insights to 8.2 billion people and to place technology at the centre of conversation for leaders navigating the future.