Google has rolled out its latest advancements in AI technology, with Veo 2 for video generation and Imagen 3 for image generation. These models are now available through Google Labs tools like VideoFX, ImageFX, and a new experimental tool called Whisk.
Veo 2: High resolution video generation
Veo 2 is a new benchmark for video generation, which generates high-quality videos over a wide range of subjects and cinematic styles. Compared to other leading models, Veo 2 stands out with its improved understanding of real-world physics, human movement, and cinematic language.
Key features of Veo 2 include:
- Support for 4K resolution and extended video lengths.
- Advanced cinematography options, such as lens-specific shots (e.g., “18mm wide angle”) and effects like shallow depth of field.
- Realistic outputs with fewer inaccuracies, such as unwanted objects or extra details.
In addition, Veo 2 considers responsible AI development by watermarking with SynthID, which is an invisible identifier to distinguish AI-generated content from other kinds of information, hence avoiding the risk of misinformation.
Veo 2 is already available via VideoFX in Google Labs, with access to YouTube Shorts and more in the coming year. Users can join the waitlist to experience Veo 2 firsthand.
Imagen 3: High quality image generation
Imagen 3 model has been improved to offer brighter, better-composed images with remarkable diversity in the art styles. The correct outputs range from photorealistic imagery through artistic styles like anime, abstract, and impressionism to better attention to prompts, textures, and finer details.
In tests with human raters, Imagen 3 achieved state-of-the-art results when compared to other leading image generation models. Starting today, Imagen 3 will roll out globally via ImageFX in over 100 countries. Users can explore the tool on Google Labs for creative image generation.
Whisk: A new way to visualize ideas
Whisk is one more creative tool from Google Labs, a combination of the visual capabilities of Gemini and Imagen 3. Whisk lets one enter or remix images over topics, styles, or scenes to produce unique visual items, like stickers or enamel pins or digital artwork, where Gemini assists in making more detailed captions based on user input, allowing these to create personalized output items.
Related
The new release of Veo 2, Imagen 3, and Whisk keeps Google ahead in the AI innovation space as it empowers creators, businesses, and daily users to bring their ideas to life. In the future, it is expected that Google Labs will have further developments of these tools through VideoFX, ImageFX, and Whisk.
