Adobe has introduced new models for image generation with Firefly

At the MAX London conference, Adobe unveiled an updated version of the Firefly platform — a unified space for generating creative content using artificial intelligence. The update introduces new models for working with images and video, as well as the ability to use models from partners, including Google Cloud and OpenAI.

Firefly now enables the creation of images, video, audio, and vector graphics from a single interface, providing a high level of control and flexibility. The platform is integrated with key Adobe products such as Photoshop, Premiere Pro, and Express, streamlining the process from concept to final product.

Key new features include:

— Firefly Image Model 4 and Firefly Image Model 4 Ultra — new image generation models featuring photorealistic detail and improved control over composition, style, and camera angles

— Firefly Video Model — a video generation model based on text descriptions and images, allowing precise adjustment of shooting parameters and animation

— Firefly Boards — a new AI-powered surface for mood boarding, conceptual development, and collaboration, available in open beta

— Firefly Services — a suite of APIs designed to accelerate creative production tasks, including the Photoshop API, Text-to-Video API, and Image-to-Video API.

The platform now supports generation using partner models. Users can switch between various generative AI tools, including OpenAI and Google Imagen, with full transparency at every stage. In the coming months, partners such as fal.ai, Ideogram, Luma, Pika, and Runway will join the ecosystem.

Firefly is available via the web; a mobile app will be released later. The Firefly Image Model 4, Firefly Image Model 4 Ultra, and Firefly Video Model are already available to all users.

Source: news.adobe.com