Google unveils Veo 2 and Imagen 3 for next-level AI creativity

Google DeepMind

@GoogleDeepMind

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

Prompt: An extreme close-up of a craftsperson's hands shaping a glowing piece of pottery on a wheel. Threads of golden, luminous energy connect the potter’s hands to the clay, swirling dynamically with their movements.

Prompt: A portrait of an Asian woman with neon green lights in the background, shallow depth of field.

5:03 PM · Dec 16, 2024

Read 279 replies