Canva releases its first image generation model
Canva is making its new image model available to all users, promising better, more customizable performance at lower costs.
TL;DR
- Canva released its first image generation model, Canva Craft, developed by its research team.
- The model works in two stages: prompt interpretation and planning, followed by image generation up to 2K resolution.
- Canva Craft focuses on creating images with accurate details and legible text, enabling precise edits.
- The model was trained using human-designed templates from Canva's library, not photographs.
- Generated images can be converted into layered designs using Canva Magic.
- Canva Craft is available to all users on the platform and via the Canva API, with future integration into third-party chatbots planned.
- The new model aims to reduce API costs, improve quality, and enhance speed control for Canva.
- Researchers from Canva's acquisition of Leonardo.ai contributed to the model's development.
- Two faster versions, Craftfast and Craft-ultrafast, are planned for release.
- Canva's valuation reportedly decreased due to a revised growth forecast.