Get all your news in one place.
100’s of premium titles.
One app.
Start reading
Digital Camera World
Digital Camera World
Gareth Bevan

Adobe unleashes its most lifelike Firefly image models yet – plus a Firefly mobile app for on-the-go gen-AI

Firefly video created using the prompt "woman withluminous dark skin and iridescent clothing looks towards the camera high fashion".

Adobe has pulled the wraps off the next generation of its Firefly generative AI, revealing a pair of new image models and an all-in-one mobile app that promises to take creators from prompt to publish on the fly.

The headline act is Firefly Image Model 4, pitched as a rapid-ideation engine for “everyday creative needs”. Adobe says version 4 can produce high-quality stills faster and with cleaner prompt fidelity than its predecessor, while remaining commercially safe thanks to a training set limited to licensed and public-domain assets.

If you need even more polish, you can step up to Firefly Image Model 4 Ultra, tuned for projects where texture, detail and complex scenes couldn't be faked – until now. I got a very brief sneak peek before the official unveiling, and the results look incredible, but I need to dig into it myself before I can make any judgments.

Video shooters aren’t left out. The new Firefly Video Model announced at the last Adobe Max is now generally available in the Firefly web app. The Firefly Video Model lets you create short clips from text prompts, images or extend existing footage – think Generative Expand for Premiere Pro.

(Image credit: Adobe)

Illustrators should also note that Text to Vector, powered by a new Adobe Vector Model, is also rolling out today for general availability. You can type your brief and Firefly outputs editable paths that slot into Illustrator – ideal for quick logo variants or social post mock-ups.

It's not all about Firefly Image Model 4, though. Adobe is also giving creators choice and flexibility to experiment with different image generation models, without having to leave the Firefly app.

These different models will provide more variety to AI generations, so you have more options for your projects. Available now are: GPT image generation, Google Imagen 3 and Veo 2, and Flux, and coming soon, fal.ai, Ideogram, Luma, Pika, and Runway.

Everything funnels into the forthcoming Firefly app for mobile. Similar to the Firefly web app, Adobe is positioning the Firefly app as a Swiss-army-like studio, with all its models for images, video, vectors, and audio generation in one place – these creations can then be imported into other apps directly on your phone.

Adobe claims more than 22 billion assets have already been generated with Firefly – a number that’s likely to skyrocket once the mobile app lands. I’ll be putting the new models through their paces as soon as they hit Creative Cloud, so stay tuned.

You might also like...

For more on the best photo editing software or the best video editing software, check out our guides.

Sign up to read this article
Read news from 100’s of titles, curated specifically for you.
Already a member? Sign in here
Related Stories
Top stories on inkl right now
One subscription that gives you access to news from hundreds of sites
Already a member? Sign in here
Our Picks
Fourteen days free
Download the app
One app. One membership.
100+ trusted global sources.