Adobe Firefly goes global and speaks your language; here's how to use it
After first launching Adobe Firefly, the “ethical” text-to-image generator, Adobe has now expanded it globally. The company says that users have created over one billion images so far, and still counting. From now on, no matter where you are, you can use Adobe’s AI image generator and create images limited only by your imagination (and Firefly’s capabilities, if we’re being honest).
[Related reading: Surprise! Adobe Stock creators aren’t thrilled with Adobe’s “ethically sourced” generative AI tool]
What’s new in Adobe Firefly
Firefly’s user base has expanded to millions of users all over the world. With AI-generated images, the company focuses on content that’s safe for commercial use. Users have generated over a billion assets so far on Firefly’s website and with Photoshop’s Generative Fill combined. According to Adobe, this makes it its most successful beta release ever.
As for the language part, Adobe Firefly supports text prompts in over 100 languages, so you can use your native language to generate images. The service will also be localized in 20 languages, and French, German, Japanese, Spanish, and Brazilian Portuguese versions are available now. Of course, I had to try it out in Serbian, but read on for the results.
“We’ve been amazed at how creators have been using Firefly to create more than a billion gorgeous images and text effects making it one of Adobe’s most successful betas ever in just over three months,” said Ely Greenfield, CTO of Digital Media at Adobe. “Today’s announcement is about making Firefly accessible to more people in their preferred languages, so they can continue to leverage our unique model to bring their imagination to life, and create the highest quality assets that are safe for commercial use.”
How to use Adobe’s text-to-image generator
Using Adobe Firefly is pretty straightforward. I find all Adobe’s tools very intuitive, and Firefly is no exception. Visit the website, and you get to choose what you want to create: an image, a text effect, generative fill or recolor, and more.
Once you choose the type of AI stuff you want to create, Firefly will give you a box for your prompt. Make it as straightforward and detailed as you can, and click “Generate” or hit Enter on your keyboard. The site will take you to a page where you can adjust your image further. You can choose an aspect ratio and apply different styles and techniques, tones, lighting, and composition. You can also change prompts and try again and again if you’re not happy with the original results.
Sample images and prompts in different languages
I played with Adobe Firefly a little to see what I’ll get. Of course, I had to test it out in Spanish and Serbian Cyrillic since the generator can understand them, so I’ll share those results, too. I tried text effects and the text-to-image generator as I’m personally the most interested in those, and you’ll see the prompts right in the box below the four resulting images. Of course, I’ll give you the English translation in the captions for the Serbian and Spanish prompts I used.
Let’s try it more detailed and in Serbian: a Dachshund wearing tiny white Adidas sneakers on all four paws Hm, maybe English will be better? No? A giraffe in a savannah, sunset, plain, twilight, silhouettes Text: Dunja (Quince); prompt: yellow quinces, red apples and green leaves a black cat playing with a red shoe a bird on a tree in a city street
The good and the bad (and the ugly)
And now, for the stuff that I liked and didn’t like about Adobe Firefly.
The good
Intuitive: as I already mentioned, Adobe’s generator is easy and intuitive. This was my first time trying it, and I found my way around in no time.
Fast: There’s no lag or bugs, everything is fast and smooth, and you quickly get the results.
It understands different languages: Firefly really does understand different languages, even mine, which is not spoken as widely as Spanish or English. I was pleasantly surprised with the text effect I got when I wrote my prompts in Serbian Cyrillic.
The results*: this is under an asterisk because not all results are good… Not even decent. But the generator surprised me by nailing the images with Spanish prompts, even though I speak Spanish at a very basic level.
The bad (and the ugly)
The results: the majority of my tests ended in a disaster, no matter how much I tried to be specific and detailed with my prompts. Some were unbearably ugly; some were just kinda weird and wonky. It was my first attempt, however, so I think I would need to practice more to get the most out of Adobe Firefly.
Potential creative and copyright issues: this is a huge topic on its own, but it’s worth mentioning. Text-to-image creations are still in the murky area of copyright laws, and they are changing the concept of creativity and art from its core. Just remember the recent case of a real photo being disqualified from a contest “for being AI.”
Final word
All in all, if text-to-image generators are your thing, Adobe Firefly can both impress you or disappoint you. I don’t think it’s there yet (like Midjourney, for example), but you can go ahead and experiment with it. You’ll find some tips and tricks on Discord, on a dedicated channel for Adobe’s AI image generator.
Source: DIY Photography
