Canva launches Craft, its first in-house image model

What happened

Canva has released Craft, its first in-house image generation model, available to all users through Dream Lab and the public API. Canva says it trained Craft from scratch mainly on millions of design templates; the model generates images at up to 2K resolution across more than 30 aspect ratios and renders text in 55 languages. Results can become editable layers through Magic Layers, while MCP integration and cheaper Craft-fast and Craft-ultrafast variants are planned.

Read next

• More changes in today’s briefingConfigure the ChatGPT desktop app with Amazon Bedrock in AWS GovCloud (US)

Reported by 3 sources

The story so far

Canva launched Canva Craft, its first in-house image generation model, making it available to all users globally through Dream Lab. The Decoder reports that Craft is also available through the public API, while TechCrunch described API access as a forthcoming feature. Canva says Craft plans layouts before generating images, supports output at up to 2K resolution, and was trained mainly on millions of human-designed templates. TechCrunch reported support for 30 aspect ratios; The Verge and The Decoder reported more than 30. The Decoder also reports that Craft renders text in 55 languages and that its results can be converted into editable layers through Magic Layers. Canva plans Craft-fast and Craft-ultrafast variants, which The Decoder describes as cheaper options. Integration through Canva’s Model Context Protocol (MCP), intended to support image generation in chatbots such as ChatGPT and Gemini, remains planned.