Adobe rolled out Firefly 4 this month, and the headline feature is Style Match: give it a reference image, and it generates new images that closely follow that image’s visual style rather than producing whatever generic look the model defaults to. It sounds like a small addition. For anyone who’s tried to keep a consistent visual identity across a batch of AI-generated images and watched every new prompt drift toward its own look, it’s the feature that’s been missing.
The update also extends Firefly Video to 60-second clips with better temporal consistency, meaning less flickering and object drift over the length of a longer generation. That matters more than it might sound: most AI video tools hold together fine for five or ten seconds and start falling apart past that, so a genuine 60-second window with consistency intact is a real capability jump, not just a longer number on a spec sheet.
The bigger shift is where this lives now. Adobe’s generative tools used to mean a round trip to a separate browser tab. Firefly’s generation is increasingly built directly into Illustrator itself, so the AI step doesn’t interrupt the actual design workflow the way bolted-on tools usually do.
None of this makes Firefly a replacement for a designer’s judgment about what “on brand” actually means. Style Match gets you closer to a reference faster. It doesn’t know why that reference looks the way it does, and it will confidently produce something technically matching and creatively wrong if you don’t check its work.
This mirrors what’s happening across creative and productivity software generally right now. Notion made a similar bet with its own AI update, building context-awareness directly into the existing workspace rather than launching something separate. For Firefly specifically, independent coverage of the visual-style matching capability is worth checking against Adobe’s own marketing language, since feature names in this space tend to shift between the official announcement and how outlets describe them.