Adobe announced Tuesday that it is introducing Firefly Image 5, the latest version of its image generation model. The company is also adding more features to the Firefly website, adding support for more third-party models and the ability to generate audio and soundtracks. Specifically, this update allows artists to use existing art to come up with their own image models.
Image 5 models can now operate at native resolutions of up to 4 megapixels. This is a significant improvement over the previous generation model, which was able to produce images at 1 megapixel natively, but then upscaled to 4 megapixels. The company says the new model also improves human rendering.
Image 5 also allows for layered prompt-based editing. The model treats various objects as layers, and you can edit the objects using prompts and tools such as resizing and rotating. The company said it makes sure that the detail and integrity of the image is not compromised when editing these layers.
Adobe’s Firefly site has supported third-party models from AI labs like OpenAI, Google, Runway, Topaz, and Flu to appeal to its creative customer base, but the company is going one step further by allowing users to create custom models based on their own art style. Currently in closed beta, this feature allows users to drag and drop assets such as images, illustrations, and sketches to create custom image models based on their own style.
The company is also adding some new features to its Firefly website, which was redesigned earlier this year. The site now allows you to use prompt boxes to toggle image or video generation, choose which AI model to use, and change the aspect ratio. The site’s home page displays your files and recent generation history, as well as shortcuts to other Adobe apps (previously in the menu).
Adobe has also redesigned its video generation and editing tools to support layer and timeline-based editing. This design change is currently only available in private beta and will eventually be rolled out to users.
Firefly also adds two new audio features. Users can now generate entire soundtracks and audio for videos using AI prompts (using Eleven Labs’ model). There are also new ways to easily create prompts. Simply select words from the word cloud and add keywords or sections.
tech crunch event
san francisco
|
October 27-29, 2025
As competitors like Canva add AI to their platforms, Adobe is catering to a new generation of creators who are increasingly using AI in their workflows. “We think of Firefly’s target audience as so-called creators and the next generation of creative professionals. We think there are emerging creatives who are GenAI-oriented. They love using GenAI in all their workloads,” Aleksandr Kostin, the company’s vice president of generative AI, told TechCrunch by phone.
He added that with Firefly, the company is freer to add new features and play around with the interface because it no longer has to stick to the muscle memory of creative professionals who are used to specific workflows in Adobe’s existing Creative Cloud tools.
