Adobe Firefly Image 5 brings support for layers, will let creators make custom models

On Tuesday, Adobe announced the launch of Firefly Image 5, the latest version of its innovative image generation model. This update introduces a host of new functionalities on the Firefly platform, including enhanced support for third-party models, and the exciting capability to generate audio elements such as speech and music. One of the standout features of Image 5 is the ability for artists to create personalized image models derived from their own artwork. The model can now produce images at impressive native resolutions of up to 4 megapixels, a significant upgrade from the previous generation, which could only generate images at 1 megapixel before upscaling them. Additionally, the new model excels at rendering human figures and introduces layered editing capabilities. This allows users to treat different elements within an image as individual layers, facilitating prompt-based edits and adjustments such as resizing and rotating. Adobe assures that these edits maintain the integrity and detail of the original image. Firefly's website has also seen enhancements, which now include features enabling users to switch between generating images and videos, select desired AI models, and adjust aspect ratios. The redesign offers a more user-friendly experience, showcasing recent work and providing direct access to other Adobe applications. Currently in a closed beta, users can drag and drop various assets—like images and sketches—to create custom models that reflect their unique styles. Adobe has also revamped its video generation and editing tools to incorporate layering and timeline editing, although this feature is yet to be widely available. Moreover, Firefly is set to launch two new audio capabilities, allowing users to generate complete soundtracks and speech for their videos with AI prompts, utilizing models from ElevenLabs. To streamline the creative process, a new feature allows users to generate prompts effortlessly by selecting keywords from a word cloud. As competitors like Canva integrate AI into their platforms, Adobe is positioning Firefly to appeal to a new generation of creators who are increasingly relying on AI tools. Alexandru Costin, VP of generative AI at Adobe, highlighted that the focus is on serving next-gen creative professionals who embrace generative AI in their workflows, providing them with the flexibility to explore new features without being constrained by traditional practices.