Adobe Unveils Firefly Video Model: A Game Changer for Creative Professionals

• Adobe has unveiled the Firefly Video Model in limited public beta, marking the company’s first commercially safe video-generating tool, as well as improvements to its current creative AI models.
• The new features, including Generative Extend and Text to Video, enable users to produce videos from text prompts and seamlessly extend clips in Premiere Pro, increasing editing productivity.
• Firefly has achieved enormous popularity, with over 13 billion photos created since its inception, and Adobe intends to keep the tools compliant and safe for business use.

At Adobe MAX, the world’s largest creativity conference, Adobe unveiled the Firefly Video Model, a game-changing addition to its family of creative generative AI models. This is an important milestone because it is the first publicly available video model built for commercial safety, broadening the possibilities of Adobe’s previous offerings, which include Image, Vector and Design models.

Since its initial beta debut in March 2023, Firefly has generated an amazing 13 billion photos, demonstrating its rapid acceptance by major brands and companies. Adobe’s new development seeks to provide creative professionals with improved tools that improve their ideation and creation processes.

New Features and Tools in the Firefly Video Model

The Firefly Video Model is now in limited public beta, with a select group of creative professionals providing feedback. This ongoing method will allow Adobe to modify and improve the model. The release of the Firefly Video Model corresponds with improvements to Adobe’s other creative tools, making it a key component of the Creative Cloud suite.

Firefly Video Model’s key offerings include:

• Generative Extend (beta) for Seamless Editing: This function, built into Premiere Pro, allows users to stretch video segments, fill gaps, and make smoother transitions. It allows users to alter film and extend sound effects and ambient noise, giving them the tools they need to make perfectly timed edits.
• Text to Video and Image to Video (Beta) for Dynamic Creations: These tools are available in the Firefly web app and allow users to create video content from text prompts or still images. Text to Video allows editors to create videos from descriptive prompts and fine-tune details such as camera angles and motion. Image-to-video enables producers to bring still photos to life, resulting in intriguing video snippets that can improve their projects.
• Firefly Image 3 Enhancements:With faster generation speeds, the most recent incarnation of the Firefly Image 3 Model enables designers to make images up to four times faster than before, reducing workflow and increasing productivity.
• Generative Workspace (beta) in Photoshop: This feature allows designers to brainstorm and iterate concepts in real-time, producing magnificent images that are more intuitive than ever before.
• Advancements in the Firefly Vector Model (beta): Adobe Illustrator has included tools such as Generative Shape Fill and Generative Recolour, which allow designers to alter vector objects rapidly and effectively.

Scaling Content Creation with Firefly Services.

In addition to the new generative AI capabilities, Adobe has developed Firefly Services, a collection of creative and generative APIs designed specifically for enterprises. The new options include:

• Dubbing and Lip Sync (beta): This technology uses generative AI to translate spoken dialogue into other languages while keeping the original voice and lip sync, hence improving video content localisation.
• Bulk Create (beta): This function enables creative professionals to edit large volumes of photos more efficiently, simplifying chores such as scaling and background removal.

These developments have piqued the interest of a variety of firms, including PepsiCo/Gatorade, IBM, Mattel, IPG Health, and Deloitte, all of which use Firefly to optimise operations and enhance creative possibilities.

A Commitment to Responsible Innovation

Adobe’s generative AI technologies are designed to promote ethical innovation. The Firefly models are trained on licensed content, so the results can be lawfully used in commercial enterprises. Adobe strives to promote trust in the creative community by upholding its AI Ethics values of accountability, responsibility, and openness.

Since the start of the Content Authenticity Initiative in 2019, Adobe has been an advocate for openness in digital content development. The program has resulted in the widespread use of Content Credentials, which serve as a “nutrition label” for digital content, indicating how it was developed and updated. This capability has now been implemented into a variety of Firefly-powered Creative Cloud capabilities.

Pricing and Availability.

The Firefly Video Model is now in limited public beta on firefly.adobe.com. Interested users can join the waitlist to gain access. During the beta phase, generations are free. More pricing and promotional information will be made available once the model is out of beta.

Adobe’s Firefly Video Model marks a substantial advancement in generative AI technology, offering creative professionals fresh tools that not only improve workflow but also ensure financial sustainability. As Adobe expands its AI capabilities, it reaffirms its commitment to supporting creatives in a dynamic digital context.

Visit http://www.adobe.com to learn more about the company and its products.