Adobe Firefly Generative AI Version 3 Looks to Catch Up to the Competition

A circular neon portal surrounded by an overgrowth of colorful coral, opening onto a serene beach scene with fluffy clouds in the sky.

At its MAX event in London, Adobe announced a huge update to its artificial intelligence system Firefly. Version 3 of the platform is significantly improved over previous iterations and promises to be able to make photorealistic images with greater variety.

After being used to generate over seven billion images over the last year (it passed a billion in its first few months), Adobe Firefly is taking a big step forward with Version 3 which enters beta today. These updates, which are available in the Firefly web application, promise massive advancement to quality and control including higher-quality image generations, better understanding of prompts, and enhanced detail and variety.

Below, Adobe shows that the same prompt can provide a range of results when specific details are absent from the text, such as varied race or ethnicity.

Portrait of a woman with shoulder-length dark hair and fair skin, looking at the camera with a slight smile, against a gray background.
Prompt: Studio portrait of young woman (output 1)
Portrait of a young woman with dark hair wearing a white and yellow lace dress, smiling subtly at the camera against a neutral-toned background.
Prompt: Studio portrait of young woman (output 2)
Portrait of a smiling asian woman with long hair, wearing a sleeveless beige top, against a grey background.
Prompt: Studio portrait of young woman (output 3)

The use of the term “photorealistic” does not appear to be an over-promise either, as generations have better and more believable lighting, positioning, detail, and the ability to display text. Until now, Firefly lagged behind its contemporaries when it came to generating images that looked like photos (or any image type at all, really). That doesn’t appear to be the case any longer.

A woman sitting on a stool in an art studio, surrounded by colorful paintings and art supplies, wearing a blue striped shirt and jeans, smiling gently as she looks to the side.
Prompt: An artist in her studio sitting at desk looking pensive with tons of paintings and ethereal atmosphere (output 1)
A young asian woman sitting thoughtfully in an art studio surrounded by vibrant floral paintings and painting supplies, resting her chin on her hand.
Prompt: An artist in her studio sitting at desk looking pensive with tons of paintings and ethereal atmosphere (output 2)
A female artist in a gray sweater sits pensively at a colorful painting workspace filled with brushes and art supplies, surrounded by vibrant paintings.
Prompt: An artist in her studio sitting at desk looking pensive with tons of paintings and ethereal atmosphere (output 3)
A golden brown cat with striking blue eyes sits on a dark backdrop, wearing a yellow polka-dot shirt, and surrounded by strands of pearls.
Prompt: Cat with pearl earrings (output 2)
A close-up portrait of a tabby cat with striking amber eyes, adorned with blue beaded hair clips on its ears, wearing a yellow ruffled collar against a black background.
Prompt: Cat with pearl earrings (output 1)

Adobe promises that Version 3 of Firefly is not only vastly improved in the quality of what it can produce but does so faster than before, too. Not only that, Firefly supposedly understands text prompts even better than before, which Adobe says allows it to create image generations that can accurately reflect long, complex prompts and include more details.

Firefly 3 brings what Adobe is calling a new style engine which allows it to deliver higher quality outputs with more variety. Outputs have a more diverse set of colors, backgrounds, and subject poses which makes them more useful across a wider range of applications.

The update adds two new features: Structure Reference and Style Reference. Structure Reference allows users to generate new images that match the general structure of a reference image, which Adobe says will eliminate the trial and error associated with writing prompts.

An elephant charges down a dusty path surrounded by lush greenery, creating a cloud of dust behind it under a clear blue sky.
Structure Reference original image
An elephant charges powerfully towards the viewer, stirring up dust under a vibrant, purple sky amidst a serene forest setting.
Structure Reference output with elephant prompt

Style Reference is what Adobe is calling Firefly’s ability to generate higher quality output, giving users more control and the ability to personalize the generations. These tools can be combined, too.

An elephant and several giraffes gather at a watering hole during a vibrant sunset, with a large baobab tree and colorful sky in the background.
Style Reference original image.
Colorful illustration of a whimsical safari scene with a sunlit sky, featuring a giraffe, an elephant, and baobab trees, stylized with vibrant, playful details.
Style Reference image output 2 to mirror colored vector image
Black and white illustration of an elephant drinking water from a pond, with a giraffe and human figure in the background under trees.
Style Reference image output to mirror black and white.

Text, which is as difficult for generative AI to produce as hands, is also improved. Adobe promises that Firefly Version 3 can make images with clear text that actually makes sense. Similarly, the platform is much better at icons, logos, raster images, and line art. All this said, Adobe did not provide any examples of text ahead of publication.

“In just over a year, Firefly has become the image generation tool of choice used by millions of creators to ideate every day, and we’re just getting started,” Ely Greenfield, chief technology officer, Digital Media at Adobe, says. “As we continue to advance the state of the art with Image 3 Foundation Model, we cannot wait to see how our creative community will push the bounds of what’s possible with this beta build.”

Close-up image of a marine invertebrate resembling an eye, with vibrant blue and orange colors, surrounded by tentacle-like structures and water droplets.
Prompt: Hyper-realistic cordyceps mushrooms in pupil (output 1)
Close-up view of a human eye showing intricate details of the iris with vibrant blue and orange colors, surrounded by dark eyelashes.
Prompt: Hyper-realistic cordyceps mushrooms in pupil (output 2)

As before, Adobe’s commitment to making generative AI ethically remains a major sticking point. The company says it develops AI in accordance with its published AI Ethics policy and Content Credentials are automatically attached to all Firefly images.

Adobe trained Firefly exclusively on licensed content such as Adobe Stock and the company has been more transparent about its training data than any other generative AI company.

A focused young man playing an acoustic guitar outdoors at sunset, with the ocean partially visible in the background. he is wearing a purple shirt and has a thoughtful expression.
Prompt: Man playing an acoustic guitar by ocean at sunset (output 1)
A man in an orange shirt plays an acoustic guitar while sitting on a wooden bench overlooking the ocean at sunset.
Prompt: Man playing an acoustic guitar by ocean at sunset (output 2)
Portrait of a young woman with light skin and blue eyes, featuring shadows of blinds cast across her face. she is looking towards the light source, creating a dramatic and thoughtful expression.
Prompt: Light leak photo of woman in black shade of light (output 1)
Portrait of a young woman with light skin and blue eyes, featuring shadows of blinds cast across her face. she is looking towards the light source, creating a dramatic and thoughtful expression
Prompt: Light leak photo of woman in black shade of light (output 2)

“[Firefly] was designed to generate content for commercial use that does not infringe on copyright and other intellectual property (IP) rights such as trademarks and logos. Adobe uses a multi-layered, continuous review and moderation approach to block and remove content that violates Adobe’s policies and offers customers IP indemnification for Firefly-generated content,” the company says.

Firefly Image 3, the official name of the latest iteration of the generative AI platform, is now available in beta and can be accessed at firefly.adobe.com.


Image credits: Adobe

Discussion