Google just released Veo 2, a new state-of-the-art AI video model. In testing, Veo beat OpenAI Sora in BOTH quality and prompt adherence. The video compilation below is 100% created by AI (more details in thread):

Veo can generate 8 second videos in up to 4K resolution (720p at launch). The model also features: — Better understanding of physics for more natural movement, lighting, etc. — Enhanced clarity and sharpness of outputs — Reduced hallucinated objects and details

The model also excels at a variety of cinematic styles, with better camera control for more creative storytelling. An animation example:

In testing, Veo beats all other top video models (Sora Turbo, Minimax, Kling, Meta Movie Gen) in head-to-head comparisons for quality and prompt adherence.

Veo 2 is rolling out now in Google Labs' VideoFX. The AI video landscape is moving faster than ever — and the video quality threshold feels like it has officially been crossed.

Alongside Veo, Google also launched Imagen 3, an upgraded state-of-the-art image generation model:

Imagen 3 excels at a variety of styles, with vastly improved text rendering and detail. some examples below of various styles and text outputs:

Like Veo, Imagen 3 also beat out all other models in testing. Rolling out in ImageFX today.

🚀 Introducing Freepik AI video generator. Everything you need to create high-quality, physically accurate videos in one place. 🤩

Looks amazing! I would love to see it on the arena board. Would it take first place? 👀

OpenAI Sora needs to get better now

Veo is incredible