Google has unveiled Veo 3.1, the latest upgrade to its Google Flow platform, introducing richer audio generation, enhanced realism, and deeper prompt alignment for creators. With this update, Google aims to make AI-assisted video editing more precise, immersive, and intuitive for professionals and storytellers alike.
Veo 3.1 now brings generated audio to Flow’s core tools – including Ingredients to Video, Frames to Video, and Extend – allowing users to synchronize visuals and sound directly within the platform. The result is a seamless creative environment where creators can refine scenes with unprecedented control and detail.
“Veo 3.1 continues to evolve based on user feedback”, Google noted, emphasizing that these experimental audio features will keep improving to match real-world creative demands.
Key Capabilities
- Ingredients to Video: Combine multiple reference images to define characters, objects, and visual style, which Flow then transforms into a cohesive, realistic scene.
- Frames to Video: Generate fluid transitions between starting and ending images, perfect for cinematic sequences or stylized motion shots.
- Extend: Create longer videos, over a minute in duration, by continuing the motion and action from the final frame – maintaining context and quality.
Veo 3.1 also brings new editing tools to Flow for fine-tuning video elements:
- Insert: Add new characters, objects, or imaginative elements while Flow automatically adjusts lighting, reflections, and shadows for natural integration.
- Remove (coming soon): Effortlessly erase objects or people while the system reconstructs the background seamlessly, preserving scene continuity.
The upgraded tools are available today in Flow, with Veo 3.1 also accessible through – Gemini API (for developers), Vertex AI (for enterprise users), and Gemini app.
Veo 3.1 forms part of Gemini API 2 and Vertex AI 3, expanding Google’s vision for multi-platform, high-quality video storytelling powered by generative AI.
