Google has unveiled its latest state-of-the-art AI models: Veo 2 for video generation and the upgraded Imagen 3 for image generation. These models are now available in VideoFX, ImageFX, and the experimental Whisk platform from Google Labs.
Veo 2 pushes the boundaries of video generation, producing ultra-high-quality videos across diverse subjects and styles. It features an improved understanding of real-world physics, human movement, and emotional expressions, delivering exceptional realism. Users can specify cinematic elements such as lens types, angles, and effects, enabling Veo 2 to create videos in resolutions up to 4K. From tracking shots to wide-angle captures, Veo 2 offers unmatched precision and creativity.
The new model includes an invisible SynthID watermark to ensure AI-generated content is easily identifiable, reducing the risk of misinformation. Veo 2 is now available on Google’s VideoFX, with plans to expand it to YouTube Shorts and other platforms in 2025.
The enhanced Imagen 3 model improves image generation by creating brighter, better-composed visuals while supporting diverse artistic styles, from photorealism to abstract art. It offers richer details, improved texture rendering, and better adherence to user prompts. Imagen 3 is now rolling out globally in ImageFX across over 100 countries.
Google’s Veo 2 and Imagen 3 represent a significant leap in generative AI, empowering creators, artists, and businesses to explore new possibilities in video and image production.
