Adobe has unveiled a new generation of generative AI Firefly models, making several updated tools for creating images, videos, and vector graphics available to users. The fourth version of the image generation model — Firefly Image Model 4 — enables the creation of images at resolutions up to two thousand pixels, with enhanced control over style, camera angles, and scaling. For more complex tasks, the Firefly Image Model 4 Ultra version is available, which handles detailed scenes and fine elements even better.
The updated Firefly Video Model is now available to all users and allows you to generate videos from text prompts or images, adjust camera movements, set starting and ending frames, and add atmospheric effects. At the same time, the Firefly Vector Model enables the creation of editable vector graphics, including logos, icons, packaging, scenes, and patterns with customizable styles.
The Firefly web app brings together all Adobe models and provides access to third-party generative models from OpenAI, Google, and Flux. Users can switch between models while creating images or videos, and every generated asset automatically receives “content credentials” — special tags with information about the author and the creation tool.
Among the new features is Firefly Boards — a space for ideas and collaboration, allowing users to create mood boards, import images, edit them, and interact with other participants. Firefly Boards is currently available in beta via the web app.
All these features are already available in the web version of Firefly, and a mobile app for iOS and Android is being prepared for launch. In addition, Adobe has opened access to the API for generating images, videos, and creating avatars, enabling the integration of generative AI capabilities into your own services and products.
