Adobe Firefly brings generative AI to mobile devices with a new app

Available for iOS and Android, the app Adobe Firefly offers creators the ability to “explore ideas, generate and edit images and videos” from anywhere by leveraging generative AI.

With Firefly mobile, creators have the freedom to generate images and videos from anywhere using prompts text (Text to image, Text to video); transform images into videos (Image to video); Add or remove objects such as people or unsightly traffic signs from your work (Generative fill), and enlarge the size of an image while filling new areas with AI-generated content (Generative expansion).

Creators will be able to explore concepts with Adobe's Firefly family of generative AI models, but they can also benefit from image generation from OpenAI, Image 3 and 4, and I See 2 and 3 from Google. Content created in the Firefly app automatically syncs with your account. Creative Cloud, making it easy to start a project in the mobile app and continue it on the web or in desktop apps like Adobe Photoshop and Adobe Premiere Pro.

Firefly: boards and compatibility with more models

Adobe has taken advantage of the launch of the Firefly app to launch the public beta of Firefly Boards, a proposal defined as “a totally new way for creative teams to explore ideas and collaborate on hundreds of concepts at the same time using multiple forms of media.” This version comes with improvements such as the incorporation of video to the canvas moodboarding powered by Firefly Dashboards AI, allowing professionals to remix uploaded video clips and generate new video sequences using Adobe Firefly's commercially safe video model or with Google's Veo 3, Luma AI's Ray2 and Pika 2.2 text-to-video.

Additionally, Adobe continues to expand its generative AI ecosystem with the integration of new partner models. While in April it added image generation from OpenAI, Imagen 3 and Veo 2 from Google and Flux 1.1 Pro from Black Forest Labs to its ecosystem, Firefly now receives the versions Flux.1 Kontext from Black Forest Labs, Ideogram 3.0 from Ideogram, Ray2 from Luma AI, 2.2 text-to-video from Pika and Gen-4 Image from Runway, as well as the latest Imagen 4 and Veo 3 models from Google.