Canva releases Craft image-generation model to all users
All Canva users can access the model through Dream Lab to generate images from written instructions.
Loading… Բեռնվում է…
All Canva users can access the model through Dream Lab to generate images from written instructions.
Canva has released Canva Craft, its first image-generation model, to all users of its design platform through the Dream Lab tab. The Decoder reported on 23 October that Craft is also available through Canva’s public developer interface, extending access beyond the design platform.
The launch adds a Canva-developed option to a platform that already offered image generation using other models. That report said Craft images could be converted into layered designs through Canva Magic. Layers let users edit separate elements of a layout.
The public application programming interface, or API, lets developers connect other software to Craft rather than use it only through Canva’s interface. TechCrunch’s 22 October launch report had described API access as forthcoming; The Decoder reported it as available on 23 October.
Canva says Craft can render text within generated images in 55 languages, according to The Decoder. That claim concerns text appearing in the image, a capability relevant to designs such as posters and menus.
Canva also says Craft placed fourth overall and first in portraits and landscapes on the DesignArena ranking platform, The Decoder reports.
Canva says Craft works in two stages. First, it interprets the user’s prompt—the written instructions—and plans where elements sit, their relative sizes, which elements stand out and how text will appear. It then generates an image at up to 2K resolution. The Decoder reports support for more than 30 aspect ratios, or width-to-height proportions.
The W27 keeps the Galaxy Z Fold 8 Ultra’s core hardware, with dark-red or dark-blue finishes, golden accents and 16GB of working memory in both storage versions.
2 min
The company says this approach produces accurate object details and readable text, and allows precise edits without changing other parts of an image.
According to Canva, Craft was trained from scratch mainly on millions of design templates rather than photographs, The Decoder reports. TechCrunch described human-designed templates from Canva’s library, including posters, social posts, presentations, infographics, menus and layouts rich in text.
Canva plans to support image generation in third-party chatbots such as ChatGPT and Gemini through its MCP connection system. Unlike access through Dream Lab and the public API, that integration remains planned for later.
The satellite operator expects access to more than twice OneWeb’s current bandwidth in return for spending in 2027–2034, with initial IRIS² services expected from 2030.
3 min
The black, gold and green edition follows September’s sold-out preorders. Nintendo’s replenished launch-day stock also quickly sold out.
1 min