TL;DR
- Announcement: Google announced three enhancements to Veo 3.1’s Ingredients to Video feature on Tuesday
- Expressiveness: Enhanced character animation from shorter, simpler prompts
- Vertical Video: Native 9:16 support eliminates cropping for social platforms like YouTube Shorts and TikTok
- 4K Upscaling: Professional-grade quality extends the tool into broadcast and commercial production workflows
- Audio: Support now spans all core generation modes, completing the audiovisual generation system
Google announced improvements to Veo 3.1’s Ingredients to Video feature on Tuesday, introducing three key enhancements for AI video generation.
The updates make videos more expressive and creative even with simple prompts, add native vertical video support in 9:16 aspect ratio for social platforms, and provide advanced upscaling to 1080p and 4K resolution. The improvements are launching across the Gemini app, YouTube, Flow, Google Vids, the Gemini API, and Vertex AI.
These enhancements build on Google’s October release, when the company first unveiled Veo 3.1 with object insertion capabilities and expanded audio support to multiple features. The latest improvements focus specifically on refining the Ingredients to Video feature, which allows users to create videos based on reference images.
Promo
Enhanced Expressiveness with Reference Images
The expressiveness improvements target a core challenge in AI video generation: producing lifelike character animation from minimal input. Ingredients to Video accepts up to three reference images of a character, object, or scene to guide the generation process.
With the update, the system produces better character expressions and movements even when users provide shorter, less detailed prompts, reducing the burden on creators who previously needed lengthy, specific descriptions to achieve quality results.
The enhancements also strengthen consistency across characters, objects, and backgrounds. Users can blend various visual elements into cohesive output without the jarring inconsistencies that plague some AI video tools.
The feature maintains character integrity across multiple shots, a critical requirement for storytelling applications where the same character needs to appear throughout a sequence.
Native Vertical Video for Social Platforms
Veo 3.1 now supports native 9:16 vertical video output, addressing a workflow friction point for creators targeting mobile platforms. Previously, users needed to generate videos in standard aspect ratios and crop them for vertical formats, often losing important visual elements.
The native vertical support eliminates this compromise, allowing creators to compose specifically for YouTube Shorts, Instagram, and TikTok from the start.
Google is integrating the feature directly into YouTube Shorts and the YouTube Create app, removing technical barriers for creators who may lack video editing expertise. The vertical format capability positions Veo 3.1 as a tool specifically optimized for the social media landscape, where vertical video dominates consumption patterns.
Professional-Grade 4K Upscaling
The addition of upscaling capabilities to 1080p and 4K resolution extends Veo 3.1’s reach into professional production environments. While AI-generated video has found traction in preview and concept work, resolution limitations have prevented broader adoption in final production.
The upscaling feature addresses this barrier, enabling creators to generate high-fidelity output suitable for commercial projects, broadcast, and cinema workflows.
The upscaling is available on Flow, the Gemini API, and Vertex AI in Google Cloud, ensuring enterprise clients and professional users have access through their preferred platforms. The 4K capability opens doors to professional applications that demand broadcast-quality resolution standards.
Audio Support Expansion
Audio integration represents a major evolution in the Ingredients to Video feature. While Veo 3, released in May 2025, introduced audio capabilities to the platform, and October’s Veo 3.1 expanded audio to other features like Frames to Video and Extend, the Ingredients to Video mode operated without sound until now. This update completes the audio rollout across the platform’s core generation modes.
Veo 3.1 generates what Google describes as richer native audio, spanning natural conversations to synchronized sound effects. Demis Hassabis, Google DeepMind CEO, characterized the industry’s progression as “emerging from the silent era of video generation.”
The audio capabilities transform Ingredients to Video from a visual-only tool into a complete audiovisual generation system, enabling more immersive and polished results without requiring separate audio production.
Platform Availability
The improvements are available across multiple access points within Google’s ecosystem. Consumer users can access the features through the Gemini app and YouTube, while professional creators have access via Flow and Google Vids.
Developers can integrate Veo 3.1 capabilities through the Gemini API, which offers both Veo 3.1 and Veo 3.1 Fast models in paid preview. Enterprise clients can leverage Vertex AI in Google Cloud for scaled deployments, with pricing unchanged from Veo 3.
Usage and Enterprise Adoption
User engagement metrics suggest strong traction for the platform. Since Flow’s launch five months ago, users have created over 275 million videos, according to Google’s Flow updates blog post from last October. This volume indicates sustained adoption beyond initial experimentation, with creators returning to the platform for repeated use.
“We’re always listening to your feedback, and we’ve heard that you want more artistic control within Flow, with increased support for audio across all features,” said Jess Gallegos and Thomas Iljic, Product Leads at Google.
This responsiveness has attracted enterprise partners seeking production-grade tools. Promise Studios uses Veo 3.1 within its MUSE Platform to enhance generative storyboarding and previsualization for director-driven storytelling at production quality. Latitude is experimenting with Veo 3.1 in its generative narrative engine to bring user-created stories to life instantly.
Competitive Context and Iteration Pace
The three-month gap between Veo 3.1’s October launch and these refinements demonstrates Google’s rapid iteration cadence in the AI video space. This pace stands in contrast to the traditional model development cycle, where major version updates arrive annually or less frequently.
The swift improvements reflect the competitive pressure in AI video generation, where Runway’s Gen-4.5 recently dethroned Veo 3 on the Video Arena leaderboard in December.
Veo 3.1 builds on Veo 3 with stronger prompt adherence and improved audiovisual quality when turning images into videos. By focusing this update specifically on the Ingredients to Video feature rather than overhauling the entire model, Google appears to be pursuing a strategy of targeted, frequent improvements over monolithic releases.
This approach allows the company to address specific user pain points and competitive gaps without the lengthy development cycles required for foundational model changes.
