Google expands its Veo AI video creation model with native support for vertical video and enhanced control via reference images.

Google’s latest release of Veo 3.1 introduces new capabilities designed to make AI-generated videos more expressive and aligned with creators’ intent. The update — highlighted by TechCrunch and corroborated by reports on broader rollout details — lets users feed up to three reference images into the model to maintain consistency of characters, objects, and backgrounds across generated content. Significantly, Veo now outputs native 9:16 vertical video formats, addressing a major demand from creators targeting mobile-first platforms like TikTok and YouTube Shorts. These improvements also include upgraded resolution options and tighter visual coherence, making AI video generation more practical for professional workflows.

Material by Iva Abadjievа

Image by Google