A new video AI generator emerges: Krea AI adds capabilities for paid subscribers
Join us in returning to NYC on June 5th to collaborate with executive leaders in exploring comprehensive methods for auditing AI models regarding bias, performance, and ethical compliance across diverse organizations. Find out how you can attend here.
When it comes to AI video generation, Open’s unreleased Sora model has so far this year stolen the spotlight away from compelling rivals Runway and Pika.
But now comes a new challenger: Krea AI, a lesser-known San Francisco generative AI media startup co-founded by Victor Pérez and Diego Rodriguez that launched its public beta in late 2023.
At the time, Krea offered AI image upscaling (adding details and improving resolution of a user-submitted image), and soon after, realtime AI image generation capabilities and a novel user interface — cursor-based brushes and simple shapes that a user could move to guide generations, in addition to more common text prompts.
Earlier today, Krea upgraded to allow one tier of its paying subscribers to generate videos. The company announced the new feature through its X account, posting a helpful thread of examples.
VB Event
The AI Impact Tour: The AI Audit
Join us as we return to NYC on June 5th to engage with top executive leaders, delving into strategies for auditing AI models to ensure fairness, optimal performance, and ethical compliance across diverse organizations. Secure your attendance for this exclusive invite-only event.
Subscribers to Krea’s priciest tier “Max,” at $60 per month (billed monthly) — the company also offers a free tier, Basic ($10/month) and Pro ($35/month) — can now visit the Krea AI website and navigate to the “Video” section, where they’ll see a new option to start creating a new video project.
How Krea Video works
Unlike Pika and Runway, which work mainly by adding motion to regions of user-uploaded still images, or converting text prompt descriptions into whole new videos, or even transforming pre-shot video into new styles, Krea presents the user with the option to use both “key frame” still images and text prompts in the same project.
Users may also select different frames to delete, re-order, or lengthen/shorten in duration.
The interface is a video timeline that will look familiar to users of more traditional video editing and filmmaking software such as Apple’s iMove, Adobe’s Premiere Pro, and even CapCut.
The enhanced video generation process introduced by Krea AI operates in two distinct phases.
Initially, a low-resolution preview is generated, giving users a quick glimpse of their creation. This is followed by a second phase that enhances the video to high resolution.
Users also have the option to cancel the generation process at any time, providing greater control and efficiency.
Already, users and backers are showing off their impressive creations using the new Krea Video feature:
While lacking the photorealism of rivals, it still is capable of generating videos with impressive smoothness, unique and captivating transitions, and eye-popping animated visuals.
A quick ascent
Perez, a former head of machine learning (ML) research at Plyzer Intelligence, and Rodriguez, a self-described “Renaissance man”, both alumni of Cornell University, co-founded Krea together in 2022. Perez now serves as its CEO and Rodriguez its CTO.
As highlighted in a recent Cerebral Valley article, Krea AI’s rapid embrace by the fast-moving AI art scene is largely due to its innovative approach to real-time AI generation.
This capability has not only set Krea AI apart from its competitors but also attracted nearly a million users, including those from the film and gaming industries.
Krea was also named one of Fast Company magazine’s “Most Innovative Companies in 2024” in March for its groundbreaking work in AI-powered media creation.
The startup is now seeking to commercialize its latest AI breakthroughs.