Google Veo 3.1 Ingredients to Video Update Adds Native Vertical Format, 4K Upscaling, and Enhanced Character Consistency

AI

by Nino LeitnerJanuary 19^th, 2026

Subscribe to News - Software

Google DeepMind has rolled out a significant update to Veo 3.1, introducing enhanced “Ingredients to Video” capabilities that transform reference images into dynamic video clips with improved consistency, native 9:16 vertical support for mobile-first content creation, and state-of-the-art upscaling to 1080p and 4K resolution for professional workflows.

The AI video generation landscape continues to evolve at a rapid pace, and Google is clearly aiming to stay at the forefront. Following the launch of Veo 3 and the Flow creative platform earlier last year, the company is now pushing its image-to-video capabilities further. These new features roll out across multiple Google platforms, including the Gemini app, YouTube Shorts, YouTube Create, Flow, Google Vids, the Gemini API, and Vertex AI.

What is “Ingredients to Video”?

For those unfamiliar with the terminology, “Ingredients to Video” refers to Veo’s capability of using reference images as the foundation for generated video clips. Rather than generating footage purely from text prompts, this approach allows creators to maintain visual consistency by feeding the AI specific images of characters, objects, backgrounds, or textures. The system then animates these elements based on accompanying text descriptions.

Google claims the updated model produces more expressive and creative results even with simple prompts, generating richer dialogue and storytelling that makes videos feel more alive. Of course, these are manufacturer claims that warrant real-world testing to verify, but the concept addresses one of the fundamental challenges in AI video generation: maintaining coherence across frames and scenes.

Character and object consistency improvements

One of the most practical improvements in this update is enhanced identity consistency for characters. According to Google, the updated Veo 3.1 model can now keep characters looking the same across multiple scenes even as settings change. For anyone attempting to use AI-generated footage in actual productions, this is crucial. The ability to maintain a consistent character appearance across different shots opens up possibilities for more complex narrative work, at least in theory.

Background and object consistency has also been improved. The system can now better maintain the integrity of settings and objects within scenes, with the ability to reuse backgrounds, objects, and textures across multiple clips. This could prove particularly useful for content creators working on series content or projects requiring visual continuity.

Additionally, the update enables seamless blending of disparate elements. Characters, objects, textures, and stylized backgrounds from different sources can be combined into cohesive clips. Google suggests using their new Nano Banana Pro image generator (Gemini 3 Pro Image) in the Gemini app or Flow to create ingredient images specifically for use with Veo 3.1.

Native vertical video and resolution options

Perhaps the most immediately practical addition for social media content creators is native 9:16 vertical video generation. For the first time, the Ingredients to Video feature can produce portrait-mode content without cropping or quality loss. Given the dominance of vertical short-form video on platforms like YouTube Shorts, TikTok, and Instagram Reels, this is a notable gap to fill.

On the resolution front, Google has introduced state-of-the-art upscaling capabilities to 1080p and 4K. The improved 1080p option is positioned as ideal for editing workflows, offering sharper and cleaner video output. The 4K option targets high-end productions and large screen presentations, capturing rich textures and detail. However, it is worth noting that 1080p and 4K upscaling is only available through Flow, the Gemini API, and Vertex AI, not through consumer-facing apps like the Gemini app or YouTube.

Availability and platform access

Google has structured availability across consumer and professional tiers. Consumers and creators can access Veo 3.1 Ingredients to Video directly within YouTube Shorts and the YouTube Create app for the first time, as well as in the Gemini app with portrait mode support. Professional and enterprise users get access through Flow, the Gemini API, Vertex AI, and Google Vids, with the higher resolution options available on Flow, the API, and Vertex AI.

The YouTube Create integration is currently rolling out to users in India, the US, Canada, New Zealand, and Australia, with broader availability including iOS planned for the coming months.

SynthID watermarking and verification

Google continues to embed its SynthID watermark in all AI-generated video, maintaining its commitment to identifying synthetic content. The company expanded its verification tool in the Gemini app in December to include video alongside existing image verification capabilities. Users can upload a video and ask whether it was generated with Google AI, a feature designed to foster transparency in an era of increasingly convincing synthetic media.

The bigger picture for filmmakers

As we have covered extensively on CineD, generative video AI represents both exciting creative possibilities and legitimate concerns for the filmmaking industry (for a great conversation about the concerns particularly, check out my extensive recent podcast talk with Hollywood DP Oren Soffer on the dangers and problems with generative AI). Tools like Veo 3.1 are undeniably impressive from a technical standpoint, but they also raise important questions about authorship, consent, and the future of traditional production workflows.

Our previous coverage of Veo 3 and Flow highlighted both the capabilities and the tensions these tools create. The partnership between Google DeepMind and Darren Aronofsky’s Primordial Soup venture continues to explore how AI-generated content might integrate with traditional filmmaking techniques, though the legal and ethical frameworks around synthetic performance and character creation remain unsettled.

For filmmakers looking to navigate this evolving landscape responsibly, our sister company MZed offers “Directing the Future: Ethical AI Video for Filmmakers” taught by Drew Geraci, known for his work on “House of Cards” and Steven Spielberg’s “West Side Story.” The course covers how to integrate AI into production workflows while maintaining creative ownership, transparency, and ethical practice. Geraci demonstrates hybrid production techniques using tools like Google Flow, combining live-action footage with AI-generated elements. Importantly, the course receives regular updates to keep pace with the rapidly changing technology, and the latest release from Google will also be covered soon.

Whether Veo 3.1’s improvements translate to genuinely useful tools for professional production remains to be seen through hands-on testing. The gap between demo reels and real-world reliability has historically been significant in AI video generation. Still, each iteration brings these tools closer to practical application, for better or worse depending on your perspective.

What do you think about Google’s continued push into AI video generation? Are features like Ingredients to Video useful additions to your creative toolkit, or do you remain skeptical about the practical application of generative video in professional work? Share your thoughts in the comments below.