TL;DR
Google updated its Veo 3.1 AI video model to generate native vertical (9:16) videos from reference images and to improve expression, motion, and consistency. The features appear across consumer and professional tools, including the Gemini app, YouTube Shorts/Create, Flow, and several cloud APIs, with new upscaling to 1080p and 4K.
What happened
On January 13, 2026, Google released an update to its Veo 3.1 video-generation model that adds native support for vertical 9:16 output when users provide reference images. The change is intended to prevent cropping for Shorts-style videos and is being added to YouTube Shorts and the YouTube Create app. Google says the model now produces more expressive character expressions and smoother movements from reference images, and it improves consistency across characters, objects and backgrounds. Users can also blend different characters, backgrounds, objects and textures to make more cohesive results. The features are available to consumers in the Gemini app, while professional access is provided via Google’s video editor Flow, the Gemini API, Vertex AI and Google Vids. Additionally, Veo 3.1’s upscaling has been improved to offer 1080p and 4K outputs through Flow, the Gemini API and Vertex AI. The first Veo 3.1 release appeared in October 2025 with audio and editing upgrades.
Why it matters
- Native 9:16 output removes the need for manual cropping or reformatting for Shorts-style social content.
- Improved expression and motion from reference images can reduce the amount of prompt engineering creators need to do.
- Availability across consumer apps and cloud APIs bridges casual creators and professional workflows.
- Higher-resolution upscaling (1080p/4K) makes generated video more usable for platforms and post-production.
Key facts
- Update announced January 13, 2026.
- Veo 3.1 can now generate native vertical videos in the 9:16 aspect ratio from reference images.
- New features are integrated into YouTube Shorts and the YouTube Create app.
- Consumers can access the features in the Gemini app.
- Professional access is available via Flow, the Gemini API, Vertex AI and Google Vids.
- Veo 3.1 produces more expressive character expressions and improved movement from reference images, even with shorter prompts.
- The model improves character, object and background consistency and allows blending of characters, objects, backgrounds and textures.
- Upscaling to 1080p and 4K is available on Flow, the Gemini API and Vertex AI.
- Veo 3.1 was first released in October 2025 with enhanced audio output and finer editing controls.
What to watch next
- Rollout timing and availability to all users beyond the listed apps (not confirmed in the source).
- How creators adopt native vertical generation within YouTube Shorts and YouTube Create.
- Pricing tiers, enterprise quotas and access limits for Gemini API, Vertex AI and Flow (not confirmed in the source).
- Content-moderation, copyright and policy handling for AI-generated vertical videos (not confirmed in the source).
Quick glossary
- Veo 3.1: A version of Google’s AI video-generation model; updated to add vertical output and other improvements.
- Reference image: An input image supplied to a generative model to guide appearance, pose, or scene elements in the output.
- 9:16 (vertical format): A portrait aspect ratio commonly used for mobile-first short-form video platforms like Shorts, TikTok and Instagram Reels.
- Upscaling: A process that increases a video's resolution, producing higher-resolution output such as 1080p or 4K from lower-resolution sources.
- Gemini API / Vertex AI: Cloud interfaces that allow developers and professional users to access Google’s AI models and services programmatically.
Reader FAQ
Can Veo 3.1 produce vertical videos without cropping?
Yes. The update adds native 9:16 vertical output to avoid cropping for Shorts-style formats.
Which Google apps and services include the new features?
Features are available in the Gemini app and are being added to YouTube Shorts and YouTube Create; professional access is offered through Flow, the Gemini API, Vertex AI and Google Vids.
Does the update improve output resolution?
Yes. Improved upscaling to 1080p and 4K is available via Flow, the Gemini API and Vertex AI.
Are pricing and full rollout details available?
Not confirmed in the source.
IN BRIEF Posted: 9:00 AM PST · January 13, 2026 IMAGE CREDITS: GOOGLE Ivan Mehta Google’s update for Veo 3.1 lets users create vertical videos through reference images Google on…
Sources
- Google’s update for Veo 3.1 lets users create vertical videos through reference images
- Google's Veo now turns portrait images into vertical AI videos
- Google Veo 3.1: What's New and How to Prep Your Workflow
- Generate videos with Veo 3.1 in Gemini API
Related posts
- Neo humanoid maker 1X unveils world model to help robots learn
- Google’s Veo 3.1 converts portrait images into native vertical AI videos
- Legion Health (YC S21) seeks founding engineer for AI-native ops infrastructure