Google continues to drive the development of AI-supported video production and presents Veo 3.1, a significantly improved version of its video generation model. The new iteration focuses on more expressive content, greater visual consistency and, for the first time, native formats for mobile platforms. At the same time, Veo 3.1 is aimed not only at creatives who like to experiment, but also at professional workflows with high demands on resolution, control and reliability.
- More expressive videos from reference images thanks to improved AI models
- Native vertical video output in 9:16 format for mobile platforms
- Upscaling to 1080p and 4K for professional production quality
- Broad integration with Gemini, YouTube, Flow, Google Vids and Vertex AI
Veo 3.1 noticeably improves “Ingredients to Video”
The centerpiece of Veo 3.1 is the enhanced “Ingredients to Video” function, which creates dynamic video clips from reference or so-called ingredient images. Google has mainly worked on the expressiveness here: Even short, simple prompts are now enough to create more vivid movements, smoother transitions and clearer visual narrative structures. As a result, the generated clips appear less static and much closer to classic video sequences.
One major improvement concerns the consistency of characters and objects. Characters retain their appearance even if scenes or backgrounds change. This allows coherent stories to be told across several clips without characters unintentionally changing their appearance. The same applies to objects, textures and backgrounds, which can be specifically reused and kept stable across scenes.
You are currently viewing a placeholder content from YouTube. To access the actual content, click the button below. Please note that doing so will share data with third-party providers.
Veo 3.1 also allows the seamless combination of different visual elements. Characters, objects, textures and stylized backgrounds can be brought together to create a unified overall image, which is particularly relevant for creative short videos and visual experiments. The improvements are clearly aimed at increasing control over the result and at the same time expanding the creative scope.
Focus on mobile and quality: vertical format and 4K upscaling
With Veo 3.1, Google is clearly responding to the trend towards mobile-first content. For the first time, “Ingredients to Video” supports native output in vertical 9:16 format. This means that videos can be produced directly for platforms such as YouTube Shorts without having to accept subsequent cropping or loss of quality. The result is full-format, full-screen clips that are specially optimized for smartphones.
At the same time, Google is improving the technical quality of the output. Veo 3.1 offers improved upscaling to 1080p and optionally to 4K resolution. While 1080p is designed for clean further processing and post-production, the 4K option is clearly aimed at high-quality productions, large displays and professional presentations. Textures, details and edges benefit visibly from the higher resolution.
The new functions will be gradually integrated into numerous Google services. In addition to the Gemini app, Veo 3.1 and the extended formats are also available in YouTube, YouTube Create, Flow, Google Vids, the Gemini API and Vertex AI. In addition, Google continues to focus on transparency: all videos created with Veo are provided with the invisible SynthID watermark and can be verified as AI-generated via the Gemini app.
Conclusion: Veo 3.1 brings AI videos closer to professional standards
With Veo 3.1, Google is taking a significant step towards mature AI video production. Improved consistency, more creative control, native high formats and high resolutions significantly expand the possible applications. The broad integration into existing Google platforms makes Veo 3.1 relevant for both short mobile clips and sophisticated video workflows. Google is not currently providing any information on prices, but many functions can be used directly via existing services.
