Introducing Adobe Firefly.
Use generative AI and simple prompts to create the highest-quality output – images, text effects, color palettes, and more.
CREATIVE GENERATIVE AI
Learn how the latest in generative AI will help you visualize any idea with AI creations.
Generative AI is an exciting technology for anyone who loves to dream and experiment with art, product design, unique text treatments and wordmarks, and more. It’s a form of artificial intelligence that uses a large language model to translate ordinary words and other inputs into extraordinary AI creations. Each word in a text prompt, like “flower growing in a junkyard,” is an instruction that tells the generative AI model what imagery to find and combine into a beautiful image.
From 3D text made of pastries to science fiction concept art, you can get inspired by all the ways generative AI can spark creativity.
Adobe Firefly is a family of creative generative AI models for Adobe products. Initially focused on both image and text effect generation, Adobe Firefly taps into a library of open-licensed images and copyright-expired public domain content to turn text descriptions into AI creations. It can create a bath bomb big bang from a few descriptive phrases, or it can add an image to text and make your brand’s name into a beautiful grassy design.
While most generative AI tools focus on a single functionality, Adobe Firefly will eventually provide a variety of tools. Firefly users will be able to remove distractions from photos, change the mood of a video, add new elements to illustrations, test out design options, add texture to 3D objects, and even create digital experiences. Current plans for Firefly include integrations into Adobe workflows that will allow you to move your AI-generated content into Creative Cloud tools like Adobe Illustrator and Photoshop.
This additional functionality with seamless end-to-end editing separates itself from other platforms like Midjourney, DALL·E, and Stable Diffusion.
Text-to-image prompt: A bee sitting on a dahlia flower in the moonlight.
It was essential for us to build generative AI models that reflect Adobe’s creator-first commitment.
To do so, the first model of Firefly trained on Adobe Stock imagery in accordance with the Stock Contributor License agreement, openly licensed content, and public domain content with an expired copyright. The goal of training Firefly like this is to avoid creating anything that infringes on the copyright of artists, photographers, and hard-working creators. In short, everything created in Firefly is designed to be commercially safe for individuals and enterprise creative teams.
You can read about our compensation model for Adobe Stock contributors whose content is used in the dataset to train Firefly models. The Firefly model does not train on Adobe Creative Cloud customers’ personal content.
By working on compensation models for creators and advocating for open standards, Adobe is creating Firefly with and for a community of makers so it improves and empowers creative work for professionals and hobbyists alike.
If you ever wanted to see what your wildest ideas look like without spending minutes — or even hours — mocking them up, Adobe Firefly is the image and photo generator for you. Perfect to help you ideate and communicate creative concepts, Adobe Firefly saves time on any and all visualizations and improve the creative process.
Whether you’re an expert designer or a small business owner, or just a curious individual, popping text prompts into Adobe Firefly can help you realize any vision, be it a dream house on the beach for a real estate brochure or the early concept art of a cute monster for a children’s book.
Text-to-image prompt: Modern New York City room with big windows, built in wooden shelving with colorful books, green leather and oak modern couch and solid oak flooring, 45mm original editorial photo.
Here are a few exciting ways you can use Adobe Firefly:
Text-to-image prompt: Driving a car on Mars.
There are many other options you can use to play with the artistic images Firefly generates. The Styles menu allows you to select from different art movements to tap into a cubist or cyberpunk style. Choose from themes that mimic everything from a product photo to concept art. You can even choose from materials like yarn or metal to adjust the texture of the image. Many of these styles can also be combined to further adjust the results.
You can also tweak the color and tone of a generated image to be warm or vibrant, select lighting that’s low or opt for a backlit shot, and even adjust the composition of the image to be a close-up or wide angle. The combinations of different adjustments give you limitless ways to create in Firefly.
In the Text Effect feature for Firefly, you can add decorative images to text. Try out sample prompts or enter a word or phrase of your choosing to get Firefly going. It will generate a version of your brand name, for example, made of driftwood or liquid gold. You’ll have the option to try other sample effects, fonts, and colors to explore every treatment before you finalize a concept you love.
The best way to learn what Firefly can do is to experience it. Explore what Firefly can help you create.
Use generative AI and simple prompts to create the highest-quality output – images, text effects, color palettes, and more.