Google has launched Veo 2, its new video generation model, and an updated version of Imagen 3, its image generation model. These tools are now available on VideoFX, ImageFX, and Google Labs’ latest experiment, Whisk.

Veo 2: Improved Video Generation
Veo 2 generates high-quality videos across different styles and subjects. Human raters compared Veo 2 to other leading models, where it achieved better results.

The model understands real-world physics, human movement, and facial expressions. Users can ask for specific effects, genres, or camera styles. Veo 2 delivers videos in resolutions up to 4K and can extend to minutes in length.

For example, users can request a low-angle tracking shot through a scene or a close-up of a scientist using a microscope. Specifying “18mm lens” generates a wide-angle shot, while adding “shallow depth of field” blurs the background and focuses on the main subject.

Veo 2 also reduces common errors seen in video models, such as extra fingers or unrelated objects.

Google prioritises safety with Veo 2. Outputs include SynthID, an invisible watermark to mark the content as AI-generated. This helps prevent misinformation and misattribution.

Veo 2 is currently available through VideoFX in Google Labs. Access will expand to YouTube Shorts and other tools next year. Interested users can join the waitlist through Google Labs.

Also Read :- Google appoints Preeti Lobana as Country Manager and Vice President for India

Imagen 3: Upgraded Image Generation
Google has also improved its Imagen 3 image generation model. It now follows prompts more accurately and supports more art styles. These include photorealism, abstract, anime, and impressionism.

Imagen 3 can create detailed textures and produce better image composition. Human raters compared it to other image generation models, where it delivered improved results.

The updated Imagen 3 will roll out globally on ImageFX through Google Labs and will be available in over 100 countries.

Whisk: New image remixing tool
Google Labs has launched Whisk, a new experiment for image generation. Whisk lets users upload or create images to mix subjects, scenes, and styles. It uses Imagen 3 along with Gemini’s image description feature. Gemini creates detailed captions for images, and Imagen 3 uses these descriptions to generate new visuals. With Whisk, users can remix elements to create items like digital plush toys, enamel pins, or stickers.

Be a part of Elets Collaborative Initiatives. Join Us for Upcoming Events and explore business opportunities. Like us on Facebook , connect with us on LinkedIn and follow us on Twitter.

"Exciting news! Elets technomedia is now on WhatsApp Channels Subscribe today by clicking the link and stay updated with the latest insights!" Click here!