Canva Releases Its First In-House Image Generation Model, Canva Craft
Understand the story
On October 23, 2026, Canva released its first in-house image generation model, Canva Craft, positioned to compete with similar image generation models from Google, OpenAI, ByteDance, and Meta. Canva Craft works in two stages: the first stage parses the prompt and generates a plan covering spatial relationships, relative proportions, visual hierarchy, and text information; the second stage generates the image based on that plan, with resolution up to 2K and support for up to 30 aspect ratios. The model emphasizes accurate object details and clear in-image text, and supports precise editing without altering other parts of the image. The above is all information disclosed in current reports; the story is still being followed.
AI overview of reporting · Updated 1 h前
Reporting timeline
Explore different aspects of the story through its reporting.
- TechCrunch · AICanva releases its first image generation model
Canva released its first in-house image generation model, Canva Craft, to compete with similar models from Google, OpenAI, ByteDance and Meta. The model works in two stages: it first parses the prompt to generate a plan covering spatial relationships, relative proportions, visual hierarchy and text information, then produces images at up to 2K resolution in up to 30 aspect ratios, emphasizing accurate object details and clear text, and supports precise editing without changing other parts.
Story heat trend
Not enough continuous observations to show a trend yet.