Adobe has launched the latest version of its AI image generation model, Firefly Image 5, introducing support for custom model creation, higher native resolution, and new audio generation tools.0
The Image 5 model now produces images at up to 4 megapixels natively, a significant upgrade from the previous 1-megapixel model. It also offers layered and prompt-based editing, allowing users to resize, rotate, and modify specific objects without losing image quality.
Adobe’s Firefly platform now lets users create personalized AI models by uploading their own art, sketches, and illustrations to train the AI on their style, a feature currently in closed beta.
Additionally, Firefly’s redesigned website enables users to switch between image and video generation, choose AI models, adjust aspect ratios, and view recent project history.
Related: Adobe Adds AI Assistants to Photoshop and Express
The update also introduces AI-powered audio generation, letting users create soundtracks and speech using ElevenLabs’ models. This marks a major step in Adobe’s effort to appeal to GenAI-oriented creators and compete with rivals like Canva.
