Google Launches Veo 2, a New AI Video Generation Tool To Beat OpenAI’s Sora

Google unveiled two AI tools named Veo2, an AI video generation tool, and Imagen 3 for image generation. To recall, OpenAI launched its Sora Turbo, a video tool powered by AI and it seems Google doesn’t want to be left behind and the tech giant introduced its new advanced AI models. The models are available via VideoFX, ImageFX, and the Google Labs experiment, Whisk. Also Read: Google DeepMind Unveils Gemini Robotics On-Device: A New AI Model That Works Without a Cloud Connection

Veo 2:

The newly launched Veo 2 promises to deliver high-quality realistic videos that would not be distinguishable from real life. The company claims it to be the state-of-the-art video generation model that can deliver high-quality output, up to 4K. Users can also explore different styles and find their extensive camera controls. The tool can carry out enhanced realism and fidelity, advanced motion capabilities, and greater camera control options. Also Read: Google is making it easy to tell AI-generated images apart

You can create a wide range of shot styles, angles, and movements with a high degree of accuracy. The model is capable of enhancing realism by understanding and having a deep knowledge of physics and human movement. It helps in generating more realistic, detailed, and lifelike videos.

Today, we’re announcing Veo 2: our state-of-the-art video generation model which produces realistic, high-quality clips from text or image prompts. 🎥

We’re also releasing an improved version of our text-to-image model, Imagen 3 – available to use in ImageFX through… pic.twitter.com/h6ejHaMUM4

— Google DeepMind (@GoogleDeepMind) December 16, 2024

The tool understands all kinds of cinematic effects, including time-lapse and aerial shots of a landscape. It has a greater understanding of language and vision along with controls for filmmaking. Additionally, the Veo 2 delivers consistency across video frames. You can also generate a video with an image along with the text prompt. The Veo 2 understands the text prompt along with the image and generates a video based on your given information. Google’s new tool can extend the videos to 60 seconds and beyond.

Google claims that Veo 2 has surpassed the other video generation tools in head-to-head comparison. The performance benchmark delivers accurate video with followed prompt, including some of the unrealistic details.

Imagen 3:

Imagen 3 is Google’s new AI image generation model, delivering realistic images, brighter, vibrant, and improved images. It can produce different styles, including photorealism, abstract, anime, and impressionism.