Adobe has launched a major update to its Firefly AI platform, including new image generation models and a redesigned web app that now features AI models from competitors. The company introduced Firefly Image Model 4, which Adobe claims offers improved quality, speed, and greater control over image outputs with resolutions up to 2K. A more advanced version, Image Model 4 Ultra, specializes in rendering complex scenes with fine details.
The redesigned Firefly web app now serves as a hub for all of Adobe’s AI models and includes third-party options from OpenAI (GPT image generation), Google (Imagen 3 and Veo 2), and Flux (Flux 1.1 Pro). Adobe plans to add more models from partners including Pika, Ideogram, fal.ai, Luma and Runway in coming months.
Other key updates
- The Firefly video model, previously in limited beta, is now available to everyone
- A new Firefly Vector Model for creating editable vector-based artwork
- Firefly Boards, a collaboration canvas for ideation and moodboarding
- New APIs, including Text-to-Image, Avatar, and a beta Text-to-Video API
- A web app called Adobe Content Authenticity for attaching credentials to work
Adobe emphasizes that all images generated through the platform will include content credentials regardless of which model is used. The company is also developing a mobile app for Firefly, though no release date was specified.