Launch the Firefly mobile app.
Last updated on
Jun 27, 2025
Learn how to generate images using the Text to image feature with Adobe or partner models like Imagen 3, Imagen 4, and GPT Image.
When generating images using the Text to image feature, you can choose from Adobe models such as Firefly Model 4, as well as partner models like Imagen 3, Imagen 4, GPT image, among others.
-
-
On the homescreen, tap Text to image.
-
In the prompt field, enter a text description of the image you want to generate.
-
Tap General > Model > Non-Adobe models to get a list of available models to choose from.
-
Tap the Add option next to Image 3.
Use the Model option to browse the Non-Adobe models available for image generation. Use the Model option to browse the Non-Adobe models available for image generation. -
Select the Aspect ratio and choose from the following options:
- Landscape (4:3)
- Portrait (3:4)
- Square (1:1)
- Widescreen (16:9)
- Vertical (9:16)
-
If you want to upload a reference image to guide Imagen 3 in generating an image aligned with your prompt, tap Subject and upload the reference image.
The subject image will be used as a reference, along with the prompt, to generate a new image. The subject image will be used as a reference, along with the prompt, to generate a new image. This subject reference image offers visual information from the reference image and the instruction from the prompt to generate a new image.
For example, if you add an astronaut image as Subject, and the prompt text is - "An astronaut standing in the lush forest."
- Preserve specific visual features, such as the exact look of your astronaut.
- Make creative or contextual edits, such as changing the background.
Image uploaded
Subject: An astronaut.
Generated image with a new prompt
Text prompt: An astronaut standing in the lush forest.
-
Tap the Generate icon to generate the image.
-
On the homescreen, tap Text to image.
-
In the prompt field, enter a text description of the image you want to generate.
-
Tap General > Model > Non-Adobe models to get a list of available models to choose from.
-
Tap the Add option next to Image 4.
-
Select the Aspect ratio and choose from the following options:
- Landscape (4:3)
- Portrait (3:4)
- Square (1:1)
- Widescreen (16:9)
- Vertical (9:16)
Choose the aspect ratio to fit where you’ll use the image, like widescreen or vertical. Choose the aspect ratio to fit where you’ll use the image, like widescreen or vertical.
-
On the homescreen, tap Text to image.
-
In the prompt field, enter a text description of the image you want to generate.
-
Tap General > Model > Non-Adobe models to get a list of available models to choose from.
-
Tap the Add option next to GPT Image.
Use the Model option to browse the Non-Adobe models available for image generation. Use the Model option to browse the Non-Adobe models available for image generation. -
Select the Aspect ratio and choose from the following options:
- Landscape (4:3)
- Portrait (3:4)
- Square (1:1)
-
If you want to use an image along with the prompt to create a modified version, tap Image and upload an image.
Use the Image option to upload an image and combine it with a prompt to generate a new modified variation. Use the Image option to upload an image and combine it with a prompt to generate a new modified variation. If you have generated an image using GPT Image earlier, the last generation is by default added here; use the x option to delete it and add a new one.
For example, if you upload an image of an astronaut and enter a prompt like 'inside a spacecraft' or 'add a mountain in the background,' the generated image will include the astronaut along with the elements described in the prompt.
Image uploaded
Generated image with a new prompt
Text prompt: Inside a spacecraft.
-
Tap the Generate icon to generate the image.
