Adobe Photoshop gets AI image generation capabilities with Firefly 3 upgrade

Adobe needs no introduction when it comes to creative tools, and it has been around for quite a while. Like any other contemporary company, Adobe too has jumped onto the AI train in a big way.

A few months ago, it introduced Project Music GenAI Control, which uses generative AI to create tunes using text. Now, Adobe’s Photoshop is next in line to get artificial intelligence capabilities that will allow users to create unique images based on prompts. 

At Adobe MAX event held in London, the company announced its latest Adobe Firefly Image 3 foundation model which will add a new image-generating feature to the platform. 

What is Adobe Firefly Image 3?

The Adobe Firefly Image 3 model is a generative AI extension for Adobe Photoshop, which can help you create images from scratch, remove them, expand content, and more. According to Adobe, the Firefly Image 3 Foundation Model was brought in for ‘massive advancements’ in quality and control. 

ALSO READ: How to edit PDFs with Adobe Acrobat

The model will also ensure higher-quality image generation along with a better understanding of prompts. This means you can expect a new level of detail, broader ranges of style, as well as variety for faster outputs and creative execution. If that’s not all, you can additionally expect better lighting, positioning, attention to detail, and even advancements in text display.

Adobe also goes on to mention that the Firefly Image 3 model was trained on licensed content. This includes Adobe Stock, along with public-domain content.

This is why, Firefly is designed to generate content for commercial use, and it does not infringe on copyright and other IP rights. Adobe states how it will offer customers IP indemnification for any of their Firefly-generated content.

How will the Firefly Image 3 model work on Adobe Photoshop?

The new addition will allow users to make the most of the generative AI tool directly using Photoshop. Firefly Image 3 will also exist as a standalone web app. That said, once you visit this website, you will be prompted to download the new Firefly 3 and integrate it into your native Photoshop app.

By doing this, you will be able to find a ‘Generate Image’ pop-up menu on the platform, where you can add your text prompt or modify from a list of existing ones, to create your image. 

How will this be different from the Generative Fill feature on Adobe Photoshop?

Adobe Photoshop already comes with a Generative Fill and Generative Expand feature, but as of today, they don’t allow users to create prompt-based images. This is set to change with the new AI-based Generate Image feature which will easily execute text-to-image generation. 

According to Adobe, the Generative Fill feature on Photoshop quickly went on to become the most used feature on the app. With the Firefly Image 3 foundation model, users will soon be able to carry out the following tasks: 

– Reference Image
– Generate Background
– Generate Similar
– Enhance Detail 

Unlike Generative Fill, which works on an existing image, you can use this new feature to create something unique or play with different styles and lighting without any additional effort.

For now, it is available with the Firefly Image 3 model on Photoshop’s beta laptop or desktop version and users may have to wait for a more large-scale roll-out.