On Monday, Google unveiled its second-generation “state-of-the-art” video-generative AI model — Veo 2, which can generate up to 4K resolution videos, along with Imagen 3, an image-generative AI model. However, neither Veo 2 nor Imagen 3 is currently available in India. This announcement comes just a few days after OpenAI unveiled Sora to its paid users.
“Introducing Veo 2, our new, state-of-the-art video model (with better understanding of real-world physics & movement, up to 4K resolution). You can join the waitlist on VideoFX. Our new and improved Imagen 3 model also achieves SOTA results, and is coming today to 100+ countries in ImageFX,” Sundar Pichai, CEO of Google and Alphabet, wrote in an X post.
According to a Google blog, “Veo 2 creates incredibly high-quality videos in a wide range of subjects and styles” and can outperform models like Sora Turbo, Meta MovieGen, Kling v1.5, and MiniMax, as rated by human judges.
Veo 2 has a better understanding of “real-world physics and the nuances of human movement and expression”, which will help the model generate hyper-realistic-looking footage with up to 4K resolution that is up to two minutes long. It can also understand cinematic jargon such as low-angle tracking shots and close-up shots and can even recreate various lens effects such as an “18mm lens” and add a shallow depth of field to the videos.
Google Veo 2 performance against other leading video AI models. (Image credit: Google Deep Mind)
With Veo 2, Google also claims that it has addressed the common hallucination effects, where most models tend to generate extra fingers or unexpected objects, to ensure the videos it generates will look more realistic when compared to the videos generated by other AI models. All the videos generated by Veo 2 will include a visible and invisible SynthID watermark to prevent the spread of misinformation and misattribution.
Veo 2 is currently available in select markets via Google Labs VideoFX and the company has plans to integrate it into products like YouTube Shorts in 2025.
The Imagen 3 by Google is a state-of-the-art image-generation tool that can create photorealistic images and is currently available in over 100 countries on Google Labs ImageFX.
Meet Whisk! 🎉 Our new experiment that lets you use images as prompts to visualize your ideas and tell your story. Try it now: https://t.co/BR1z7gmDs6 pic.twitter.com/2zrPLQZlga
— labs.google (@labsdotgoogle) December 16, 2024
Besides these two announcements, Google is also expanding the availability of Whisk, a Google Labs experiment that lets users create images, refine, and remix, which is now powered by Imagen 3. Unlike most image-generative AI models, Whisk does not require complex context, and the same is automatically generated using Gemini.
Why should you buy our Subscription?
You want to be the smartest in the room.
You want access to our award-winning journalism.
You don’t want to be misled and misinformed.
Choose your subscription package
