Adobe Firefly Image 3 Model: Experience Enhanced Photorealism

Just one year after launching Adobe Firefly, Adobe is thrilled to introduce their newest image generation model, Adobe Firefly Image 3 Model, with stunning advancements in photorealistic quality, styling capabilities, detail and accuracy, along with greater variety.

The new Adobe Firefly Image 3 Model — available in beta in the Firefly web application, Adobe Photoshop, and in Adobe InDesign — brings new levels of creative expression and control to all. Let’s take a look at what’s specifically new in the Firefly web application.

Introducing the Image 3 Model in Text to Image on the Firefly web application

Now, you can try out the Image 3 Model in the Text to Image module on the Firefly web application. With this update, Adobe is making it faster and easier for you to realise your creative vision and tell your story. And just like the first two Firefly image models, Firefly Image 3 was designed to generate content that’s safe to use in your commercial work.

Let Adobe know what you think by clicking the “Rate this result” button available on every generated image. You can also share feedback on Discord, Adobe Community or help compare outputs in the Rate Firefly images section of the website. Adobe couldn’t build Firefly without their community and your feedback helps them grow and improve.

Here are just a few examples of how the innovations in Firefly Image 3 model benefits you in your day-to-day work:

• Enjoy new levels of creative expression: The new model produces higher quality images, better interpretation of prompts, automatic application of styles that match your prompt, and more accurate text in images, allowing creatives of all skill levels to ideate easily.
• Gain enhanced creative control in the ideation process: Image 3 Model also works with our recently released Structure Reference and Style Reference capabilities to deliver exceptional user control and state-of-the-art visual quality. In addition, the new model has an improved ability to take reference images into account with a prompt. Eliminate the trial and error of having to write the perfect prompt by using reference images.
• Achieve a creative vision faster: Image 3 Model has a better understanding of text prompts and scenes, enabling image generations that better reflect long, complex prompts and include richer details, including text.

Also, Adobe is introducing the Generative Expand feature in the Generative Fill module on the Firefly web application. This opens up creative options by enabling you to change the aspect ratio or size of an original image. Now, you can extend images to fit any content needs like making space for text or fitting an image to a size specification.

Adobe’s commitment to responsible AI

Firefly has been used to generate over 7 billion images worldwide since its initial beta release in March 2023. Adobe built it for direct integration into workflows Adobe customers use every day, including Adobe Photoshop, Adobe Express, Adobe Illustrator, and Adobe Substance 3D. In just one year, Firefly has transformed image editing, template creation, vector design, and 3D texturing and staging for the better.

A technology as powerful and transformative as AI requires a thoughtful approach, though. That’s why Adobe remains committed to developing generative AI responsibly with creators at the centre. Adobe co-founded a global coalition called the Content Authenticity Initiative (CAI), which now has more than 2,500 members from across industries working together to promote transparency in digital content with Content Credentials.

In particular, Adobe attaches Content Credentials to assets produced using Firefly so that people who work with or view the content can see how it was made and whether AI was involved. This ensures we establish a transparent chain of trust from creation to dissemination online.