Canva launches Craft, its own image model that plans the layout first, with readable text and parts you can edit

Canva released Craft on 22 October, the first image model built and trained by its own research team. It works in two steps: it first plans where things go, how big they are and what the text says, then draws the image at up to 2K resolution in more than 30 aspect ratios. Canva says it is available now in the app and through its public API, and was trained on design layouts rather than photographs.

• Open Canva and try Craft on something you need this week, such as a menu header, a sale poster or a social post, then read every word of the text before you use it. Check what your plan allows.
• Use Magic Layers on the result to turn it into separate editable pieces, so you change a price or a date yourself instead of generating the whole image again.

Source: Canva. This is our short summary of their reporting. The paragraphs marked "Our read" and the things to do are SeedFoundry's opinion. The summary is drafted with AI tools. Image: Canva, from its own announcement. How we source and credit