Discover new ways to ideate and create assets with the Text to Image feature (‘Generate Image’ in Ps UI), powered by the new Adobe Firefly Image 3 Model. Use generated assets to create custom composites without leaving Photoshop or combine multiple generated images on a canvas to ideate and refine ideas more efficiently.
This update gives you even greater control over your generative workflows and outputs, unleashing more freedom and flexibility than ever before for personal use, work projects, or anything in between.
Control and guide the Text to Image (‘Generate Image’ in Ps UI) feature to ideate or create the exact images you want using additional settings for Reference Image (e.g., allows you to upload an image with a desired style aesthetic that will influence the generated result), Content Type (i.e., photo or art), and Style Effects (e.g., Movements, Themes, Techniques, Effects, Materials, Concepts).
Photoshop AI image generator, powered by the Adobe Firefly Image 3 Model, creates images from text prompts using various creative controls within the app. Firefly is Adobe's broader generative AI platform used across multiple Adobe applications.
Generative AI features in the Photoshop desktop app are powered by the commercially safe Adobe Firefly Image 3 Model. This ensures you can push the bounds of your creativity, knowing that the current Firefly generative AI model is trained on a dataset of Adobe Stock along with openly licensed work and public domain content. Please note that generative AI features in the Photoshop (beta) app are not for commercial use.
Generate Image empowers you to ideate and create the exact assets you want using additional settings for Reference Image. You can also explore the Firefly Community Gallery with user-generated images and prompts for inspiration or editing.
In contrast, Generative Fill adds or removes content from a selected area of an existing image, ensuring the new content matches the lighting, shadows, and perspective for seamless blending without needing detailed text descriptions.