Adobe has introduced two new versions of its Firefly text-to-image AI model, along with several updates to its Creative Cloud apps, including Photoshop and Illustrator.
The new Firefly models aim to make image generation faster, more realistic, and more customizable.
The fourth-generation Firefly Image Model is now available and is designed to deliver images in up to 2K resolution. Adobe says this is its fastest and most controllable model yet. Users can expect better control over image styles, formats, and camera angles. For more advanced needs, Adobe has also launched Firefly Image Model 4 Ultra, which is better at handling detailed and complex scenes.
These new AI tools are available on Adobe’s Firefly web app. Adobe’s text-to-video and text-to-vector tools, which were previously in beta, are also now available on the platform. A new tool called Firefly Boards, first shown during Adobe’s Max event last year, is now in public beta. It works like a collaborative moodboarding app, similar to FigJam. Adobe also confirmed that a Firefly mobile app is coming soon for both iOS and Android devices.
Adobe is also opening its platform to third-party AI models. Users can now generate images using OpenAI’s image model, Google’s Imagen 3, and video content using Google’s Veo 2. Support for models from Luma, Pika, Runway, fal.ai, and Ideogram is expected soon. However, Adobe marks its models as “commercially safe” since they’re trained on licensed or public content, unlike some third-party tools.
In addition to Firefly, Adobe has updated its Creative Cloud apps. Illustrator now includes AI-powered shape fills and text-to-pattern tools, which were in beta last year. Photoshop users will see improved tools for color adjustments and automatic selections for things like hair, clothes, and facial features. The Actions panel in Photoshop has also been updated to offer editing suggestions based on a user’s unique style. Adobe says this feature will be part of a larger creative AI assistant coming in the future.
