Google adds vertical video creation to the Veo 3.1 AI model

Credit: Google

Google released updates to its Veo 3.1 artificial intelligence video generation model on January 13, 2026, introducing native vertical video outputs in a 9:16 aspect ratio for its Ingredients to Video feature, along with upscaling capabilities to 1080p and 4K resolutions.

The enhancements allow users to create full-screen portrait videos from reference images without cropping, targeting platforms such as YouTube Shorts.

The Ingredients to Video tool, which generates clips based on up to three reference images, now produces more expressive content with consistent characters, backgrounds and objects across scenes.

Users can blend elements from different images, including characters, textures and stylized settings, into cohesive videos. The model supports eight-second clips at improved 1080p for sharper quality and 4K for detailed textures suitable for professional editing.

These features became available in the Gemini app, YouTube Shorts, the YouTube Create app, Flow video editor, Google Vids, the Gemini API and Vertex AI starting on the announcement date.

"Our latest Veo update generates lively, dynamic clips that feel natural and engaging — and supports vertical video generation," said Ricky Wong, lead product manager at Google DeepMind.

Videos created with the updated tool include a SynthID digital watermark for verification, and users can check AI-generated status through a tool in the Gemini app.

The vertical support extends previous text-based prompt capabilities introduced in 2025, now applying to image references for mobile-first content creation.

"For the first time, 'Ingredients to Video' supports generating videos in a native 9:16 aspect ratio," according to the announcement.

Upscaling to higher resolutions applies across all Veo models in select platforms.