Be Tech Ready!!
AppsArtificial Intelligence

Adobe dives deeper into generative AI with new Firefly models; to make human images more realistic


At the Adobe Max conference, Adobe announced that it has updated the models powering its generative AI image creation service, Firefly.

The company has said that the Firefly Image 2 Model will be better at rendering humans, including facial features, skin, body, and most importantly, hands.

The generative AI text-to-image service has generated three billion images since Adobe launched it about six months ago, out of which one billion images were generated in the last month alone.

As per the company, the Adobe Firefly Image 2 Model advances Firefly’s creative control and image quality, Firefly Vector Model, which is said to be the world’s first generative AI model for vector graphics, now powers the new Text to Vector Graphics in Illustrator and the Firefly Design Model instantly generates template designs.

“Our rapid, continuous pace of Firefly innovation continues with imaging, design, and vector generation advancements – three new models that set new industry standards for output quality and user control,” said Ely Greenfield, chief technology officer of Digital Media at Adobe.

The upgrade to the model now brings advancement to creative control and quality, including new text-to-image capabilities to the next generation of generative AI.

Adobe has introduced three main upgrades: Generative Match to create content in user-specified styles; Photo Settings to enable photography-style image adjustments; and Prompt Guidance which allows users to reword their prompts for better results.

The upgrade also introduces a few new controls in the Firefly web app. Users will now be able to set the depth of field for their images. Motion blur and field of view settings have also been made a part of the web app.

Another interesting feature is the ability to upload an existing image and have Firefly match the style of that image. A new auto-complete feature will help complete prompts to create better images.

Currently, the new model is available through the Firefly web app as a beta. It will soon come to Creative Cloud apps like Photoshop, Illustrator, Lightroom, and Premiere Pro in the near future.

While many of the new features will be available in the Adobe Express all-in-one editor, which is available for free and paid tiers, the non-paying members can use a certain number of credits every month to try the AI features.