ByteDance launched Goku, an advanced AI model for image and video generation. The Chinese tech company's latest AI models can generate realistic videos of people interacting with products, potentially transforming how companies create advertising content. However, it also lowers the barrier for businesses to create value.

Goku produces high-quality visuals from text prompts. It uses rectified flow transformers, a type of AI model that generates images and videos more smoothly and accurately by improving the way information flows through the model. Think of it like a high-tech artist that refines its brushstrokes to make digital creations look more natural and realistic.

Bruce Burke Participating In Forbes Entrepreneur Of Impact Competition

VOTING IS NOW OPEN - I HAVE TO ASK - PLEASE VOTE FOR BRUCE BURKE

I’m currently in 13th place, but have to get into the Top 10, by THURSDAY the 13th!

Exciting News! Neural News Network editor Bruce Burke in the top 15 participants in the Entrepreneur of Impact competition. One visionary winner will be featured in Forbes, receive $25,000, and have a one-on-one mentoring session with the Shark Tank's own Daymond John. Please vote for your Neural News Network editor!

CLICK HERE TO VOTE FOR BRUCE BURKE IN ENTREPRENEUR OF IMPACT

I'm proposing building an AI-powered, fully automated news and information organization that creates news articles, videos, podcasts, deep dives, special reports, white papers, and more — focused on the ever-expanding world of AI.

I’m currently in 13th place, but have to get into the Top 10, by THURSDAY the 13th.

I would appreciate your vote and will be posting again when voting starts. I have setup my profile that outlines my proposal linked below. PLEASE VOTE TODAY!

CLICK HERE TO VOTE FOR BRUCE BURKE IN ENTREPRENEUR OF IMPACT

Goku models can generate realistic product videos without human actors. ByteDance built its new Goku AI models using a massive dataset of about 160 million image-text pairs and 36 million video-text pairs, according to the accompanying paper. The data comes from academic datasets, internet sources, and partner organizations. Unlike other video models, Goku can create both still images and videos from text descriptions. The system uses a new transformer architecture with 2 to 8 billion parameters that handles both formats simultaneously.

In benchmarks, Goku performs well in both image and video generation. The video model, Goku-T2V, scored 84.85 on VBench, surpassing similar tools from companies like Kling and Pika. The output quality also shows clear improvements over ByteDance's previous Jimeng AI model.

ByteDance has shared several sample clips on the project page, ranging from realistic to creative scenarios. While the company hasn't specified Goku's limitations, the examples are all four-second clips at 24 FPS in 720p resolution.

More on Goku Open Source video generation tools

Please support our podcast sponsor, Moving On IT | Tomorrow’s Technology Today

Moving On IT | Authorized Partner For IT, AI, And Cybersecurity Solutions

I’ve partnered with Moving On IT, your authorized partner for navigating the complex landscape of today’s technology. Moving On IT specializes in providing cutting-edge hardware, software, and cybersecurity solutions tailored to your needs.

From robust IT infrastructure to advanced Al applications, Moving On IT empowers businesses to thrive in the digital age. Contact Moving on IT with all your IT, AI and Cybersecurity requirements. Call +1 (727) 490-9418, or email: [email protected]

Check out Moving On IT’s new press release on Cybersecurity Dive | CLICK HERE

Get full access to Neural News Network at remunerationlabs.substack.com/subscribe