Generate images from text in Adobe Firefly
Turn simple prompts into stunning visuals using settings, references, and style controls to shape your results.
Published
Turn your words into stunning visuals with just a click.
Welcome to the magic of Text to image in Adobe Firefly.
I'm Jesús Ramirez.
In the next few minutes, I'll show you how to create high-impact promotional graphics for a guitar instructor, using nothing but a simple text prompt in Adobe Firefly.
From the Firefly website, at firefly.adobe.com, click on Text to image.
From here, you can enter a prompt in this text box.
To generate great images, write clear and descriptive prompts that describe what you want to see.
Describe the subject, add important details, and include the setting or location.
No need for command words like create, make, or imagine.
We'll begin with a concise and specific description of the subject, 'man playing a guitar'.
Before generating, let's set some of the General settings.
Choose the Model you like.
Adobe is always releasing newer models with faster performance and more nuanced results.
Just select the latest one from this dropdown.
Then set your desired Aspect ratio.
I'll choose Landscape (4:3) to generate a horizontal image.
I'll also choose Photo as the Content type to generate a photorealistic image instead of an illustration.
Then I'll click the blue Generate button.
In a few moments, Firefly will generate four variations based on my prompt.
You can click on them to expand them into a full screen mode to get a better look, and click on the arrow icons to cycle through the results.
To exit the full screen mode, click on the Close button.
The results look good, but they're generic.
Let's try a more descriptive prompt with more details about the main subject, actions, and location: 'young man with a beard and long brown hair, wearing a black leather jacket, playing an acoustic guitar on a rooftop with a cityscape behind him'.
And I'll click the blue Generate button again.
There is no need to use complete sentences.
Firefly works great with simple phrases, and you can use commas and semicolons to keep text prompts organized and legible.
And here we have our four new generations.
They are much closer to what I had in mind, but we're still missing the mood that I want for this image.
To invoke a mood, we can add more emotion to the prompt.
In this case, I will add 'joyful smiling' to the start of the prompt, and I'll generate by tapping on the Enter key on Windows - that's the Return key on the Mac.
This prompt offers a clear subject, a distinct mood, strong visual cues, and a location, making the generated image more closely resemble what I had in mind.
And as amazing as these generations are, they're not quite there yet.
I could improve my generation by adding more to the prompt, but longer prompts aren't always better.
It's a good idea to use the friendly controls on the left to give Firefly more direction, helping you get closer to your vision, or even explore new ideas.
I'll scroll down to see two key controls: Composition Reference, which influences how the visual elements and subjects are arranged within the frame, and the Style Reference, which influences the look and feel of your generated images.
You can use the provided images or upload your own for better customization.
In this case, let's upload our own Composition Reference.
I want to make sure we have the appropriate negative space on the left side of the image in the rough scale of the guitar player.
I drew this graphic in Adobe Photoshop, and you can download it for free to try in your generations.
To upload a Composition Reference, click on the Upload image button and find it on your device.
You'll see the Reference photo in this thumbnail.
Also, keep in mind that we have access to the Strength slider, which controls the adherence level to the reference sketch we uploaded.
At about the center works in this case.
And I'll Generate.
Now all the generations closely resemble my Composition Reference, which is exactly what I want.
You can continue adding Effects and Styles to create better results.
Under the Effects settings, scroll down to the three dropdowns.
I will set the Color and tone dropdown to Warm tone.
Under Lighting, I will choose Dramatic light.
And under Camera angle, I will select Shallow depth of field to make sure the background is blurry.
Then I'll Generate.
Those little details both in your prompt, your Reference image, and your chosen Styles help Firefly generate better images that much more closely resemble your ideas.
I really like how these generations turned out, but I want to give my editor more options.
Let's try a more stylized version just in case.
I'll remove the Shallow depth of field, Dramatic light, and Warm tone effects by clicking on their corresponding X icons.
Then from the Movements group, I'll choose Film noir for a more moody vibe, and I'll Generate.
The results are fantastic and exactly what I want.
Now that you know how to generate images in Adobe Firefly, try turning your wildest ideas into a reality with the Text to image feature.
Again, my name is Jesús Ramirez.
Thank you for watching.
What you’ll learn
Write descriptive prompts
Use short, detailed phrases that describe subject, action, setting, and mood — no need for full sentences. Use automatic prompting for expressive results or check out expert tips in Writing effective text prompts.
Use references for structure and style
Upload your own sketch or choose built-in examples to guide layout and visual style.
Fine-tune with effects and presets
Adjust lighting, color tone, and camera angle to better match your creative vision.
Instruction by
