Business demands full dedication—success comes from competing and pushing boundaries.
Since the launch of ChatGPT by OpenAI, the company ignited a fierce competition in the tech world. Adobe is among the giants that answered the call, and its product is called Firefly.
It was originally an AI image generator, meant to enhance its existing products.
But since others have ventured further, and into creating AI video generators, Adobe doesn't want to be left to far behind.
This is where it came up with what it calls the 'Firefly Video Model.'
What it does, is rival OpenAI Sora and some other big names, but the Adobe's way.
The ultimate in creative AI. Now with video. #AdobeFirefly Generate Video gives you unparalleled creative control and IP-friendly outputs.
Check out what's new with Firefly and get ready to create like never before. https://t.co/phnOwQj8RX pic.twitter.com/jv41EfM7ib— Adobe (@Adobe) February 12, 2025
In a blog post, Adobe said that:
Just like others in the competition, Firefly Video Model, which was first announced back in October 2024, allows users to create videos from text prompts or images. Users can also specify camera angles to control shots, craft unique atmospheric elements, and develop custom motion design elements.
With it, users can create videos, translate audio into several new languages, and create a video clip with exactly the style they want.
Initially in beta, the idea is to build Firefly for multi-modal creation, "bringing together video, image, and vector generation all in a single application to streamline your creative flow."
Videos generated by this Firefly Video Model are capped at 1080p, at 24 frames per second, up from the original 720p quality.
Both the Text-to-Video and Image-to-Video features take 90 seconds or longer to generate clips at a maximum length of five seconds.
This, apparently, is shorter than the 20-second duration available to Sora users.
Adobe said that it's working on creating a both a faster, lower-resolution "ideation model" and a 4K model, which are "coming soon."
"We actually think that great motion, great structure, great definition scheme, making the actual clip look like it was film, is more important than making a longer clip that's unusable," said Alexandru Costin, Adobe's vice president of generative AI.
But where it lacks, is complemented with the fact that Adobe is already a digital media software juggernaut, meaning that the company can integrate this model to its existing array of products.
According to Adobe, Firefly Video integrates with Adobe Creative Cloud editing and workflow tools.
In another post on its website, Adobe said that "the Firefly app is the all-in-one professional destination that inspires users to seamlessly ideate and create production quality work with unparalleled creative control, multi-modal workflows and integration with industry-leading Creative Cloud applications."
The new Firefly Video Model— the industry’s first commercially safe AI video generation model — powers Generate Video (beta) in the Firefly application, as well as Generative Extend (beta) in Adobe Premiere Pro, and generates IP-friendly video content that can be used in production today."
This is where Adobe pushes its subscription-based business to those hardcore users who wish to utilize the model to the max.
Adobe introduced two new offerings — Firefly Standard and Firefly Pro — that give customers access to premium Firefly video and audio features. All Firefly plans include unlimited access to Firefly imaging and vector features and tiered capacity for premium video and audio features so customers can choose the right capacity for their generative AI needs.
Other big names that have entered the AI video generator competition, include ByteDance with OmniHuman-1 And Goku, and also with Loopy, Alibaba with MIMO and also EMO, Runway with Gen-3 Alpha and Act-One, Tencent with DynamiCrafter, among others.