Veo 2

Experience unparalleled realism and control in AI-generated videos with Veo 2

Veo 2, developed by Google DeepMind, is a state-of-the-art AI video generation model that transforms text prompts into high-quality, realistic videos. Building upon its predecessor, Veo 2 offers enhanced resolution, extended video duration, and improved physics modeling, positioning it as a formidable competitor in the AI video generation landscape.

When Google first released examples of Veo 2’s capabilities it was immediately clear to most observers that the AI video generator market had a new benchmark. Veo’s videos were more naturalistic and less immediately uncanny than we’d seen before – it seems to have a better handle on the physics of movement and delivers impressively detailed image quality. The fact that it also allows you to generate considerably longer videos than the competition – up to 2 minutes – is also significant.

This impression is backed up by human testing. If Google’s research is to be believed, its video generator was the clear victor when test participants viewed the results of over 1,000 prompts. Veo 2’s generations were preferred to Sora Turbo’s 58.8% of the time, compared to 14.5% who were neutral, and 26.7% who preferred Sora Turbo.

Needless to to say, AI video tech is rapidly evolving and the landscape could look very different in a matter of months, but going into 2025 it’s hard to dispute Veo 2’s status as the leader of the text-to-video pack.