Google Veo 2 and Imagen 3: the response to Sora Turbo with 4K video and improved physics
In one sentence Google DeepMind announces Veo 2, a text-to-video model with up to 4K output and 2-minute clips, and updates Imagen 3 — released on VideoFX/ImageFX and later in the Gemini app stack.
Early December 2024 OpenAI finally launches Sora Turbo, after a year of anticipation. Days later Google responds by announcing Veo 2, its video generator, and Imagen 3, its image generator.
Veo 2 promises output up to 4K resolution and clips up to 2 minutes — significantly more than Sora (1080p, 20 seconds). It also claims better real-world physics: objects fall realistically, fluids behave like fluids, reflections are correct.
Access rolls out gradually, first via VideoFX (Google's experimental lab), then integrated into YouTube Shorts and the Gemini app in 2025.
Companies
Google, Google DeepMind
Tools
Veo 2, Imagen 3
Tags
Sources