Adobe Unveils Firefly V2, With Ability To Copy & Paste Styles For Continuity

Image courtesy of Adobe

Adobe Firefly has been an undisputed game-changer for users of all skill levels, empowering beginners to conjure up visuals without having to wield a paintbrush while giving creatives and businesses peace of mind with its copyright-safe nature. Now, the “ethical” AI art-generating model has been given more wings. At this year’s Adobe MAX, the tech giant introduced the Firefly Image 2 Model, which aims to take the creative potential of Firefly to new heights.

The Firefly Image 2 Model boasts a larger model capacity and improved image generation capabilities, making it quicker and simpler for users to bring their creative concepts to fruition, according to Adobe. You can now test its capabilities in beta on the dedicated Firefly web app, where you can also compare the differences between the old and new versions.

Image courtesy of Adobe

Generative Match

Image via Adobe

The biggest addition, besides enhanced image quality, is the new ‘Generative Match’ feature, which allows users to apply a style from a pre-selected set of images or upload their own style reference image. This makes it easy to create visuals that align with a specific style or brand, facilitating the creation of consistent content across different assets.

By combining text prompts with reference images, Firefly’s generative AI produces imagery that matches both, ensuring consistency in style and feel. For example, if your reference image features a whimsical cartoon-style cat, your generated image will adopt the same style.

Image generated on Adobe Firefly, via Adobe Blog

Adobe notes that this feature is particularly useful for artists looking to maintain their unique style or marketing teams aiming for a consistent look across various assets. It can accelerate content creation while providing a creative head start.

To give Generative Match a go, head over to the Firefly web app. Under the Text-to-Image feature, look for ‘Style Match’ on the right. From there, you can choose from a set of preloaded reference images or upload your own image. Do make sure you have permission to use the artwork if you choose the latter option, however.

Video via Adobe

Generative Match technology is also now available in Adobe Illustrator, with broader access coming to other Creative Cloud apps in the near future.

Photo Settings to AI-Generated Works

Users can now apply and fine-tune photo settings, similar to manual camera controls, to Firefly outputs. This results in images of higher photorealistic quality with enhanced details, including features like skin pores and foliage, as well as effects such as motion blur and depth of field.

Prompt Suggestions

Firefly now assists in generating superior results by suggesting prompt expansions or rewordings. This option streamlines the creative process by proposing descriptions designed to produce stunning outputs.

Image generated on Adobe Firefly, via Adobe Blog

Sharing

Finally, users can now share links to Firefly-generated images and their associated prompts directly from the platform for collaborative purposes. Additionally, it’s now possible to save generated content to Creative Cloud Libraries for easy access within other Creative Cloud applications.

Since its beta launch in March, Firefly has been instrumental in the creation of over three billion images. Importantly, Firefly is built upon a foundation of Adobe Stock, openly-licensed, and public-domain content, ensuring that creator rights are respected.

Adobe recently announced that Firefly outputs can now be used commercially. “This includes any type of usage that generates commercial activity for you, your brand, or your business,” reiterates the company.

[via Adobe]