Learn more about what’s possible with Firefly and why you might encounter certain limitations.
Understand known limitations in Firefly so that you use it more effectively and make informed decisions when accessing features, such as Text to image or Generative fill.
Firefly access
Unable to use Firefly features with Shared Device Licenses
Shared Device Licenses only include access to desktop apps, not services such as Firefly. To use Adobe Firefly generative AI features, contact your IT admin and request a named user license or an Adobe Express license.
Prompts
Prompts aren’t accurately translated
We currently support prompts in over 100 languages using machine translation to English provided by Microsoft Translator. Because of each language's nuances, some generations based on translated prompts may be inaccurate or unexpected. The Firefly team is working hard to identify and resolve any issues as we find them. To help with this, you can report unsuccessful translation results using the Report icon available in every image.
Can't enter a prompt and change the style without refreshing images
When generating image variations in the Firefly website, using a new prompt or adding modifications to the existing text prompt requires generating new images. However, we are actively developing the capability to type a new prompt and adjust the theme and style without the need for image refresh.
Generated results
Distorted text in generated images
We recognize some irregularities in certain scenarios, but text and symbol generation in images still needs support in the Text to Image feature. You can report images with distorted or unclear text using the Report icon, which appears when you hover over a generated image.
Unrealistic body features in generated images
The quality of our models will improve with time, and generated content will continue to better follow prompts. Select the Report icon to provide feedback about distorted facial or body features.
Firefly doesn’t generate content featuring famous people or brands
Firefly only generates images of public figures available for commercial use on the Stock website, excluding editorial content. If your text prompt includes a famous person, company, or branded item not available in Adobe Stock, Firefly won’t be able to use this information to generate an image. You’ll need to revise your prompt to exclude those references.
Favorited images
Styles aren’t getting saved with favorited images
Favorited images that use reference images or those generated with the Show similar feature aren’t fully optimized yet and may produce inconsistent results. Additionally, very long prompts might lose applied Styles when saved. We’re actively working on resolving these issues.
Favorited generated images are deleted
Favorites are saved using your browser data so they won't be available on another browser, in incognito mode, or if you clear your browser data. So, ensure that you are accessing the Firefly website using the same browser that you used while saving the favorites.
To access your images generated from the Text to image on other browsers and devices, save the images to Creative Cloud Libraries.
More like this
Join our community to connect, learn, and engage
For inspiration, expert tips, and solutions to common issues, visit Discord or the Adobe Firefly Community forum. Connect with our team and fellow users to exchange ideas, share your creations, stay updated with the latest features and announcements, and provide feedback.
