Google DeepMind and Google Labs revealed a suite of updates to its AI video tool Flow, introducing Veo 3.1 with enhanced creative controls, richer audio capabilities, and improved narrative realism.
Flow, Google’s AI filmmaking platform powered by Veo, has enabled users to generate over 275 million videos to date. The new features deliver more granular control over visual storytelling and bring audio support across core functionalities that were previously silent.
Veo 3.1 augments the existing Veo 3 model with stronger prompt fidelity, higher audiovisual quality, and more believable textures. The update introduces generated audio support for “Ingredients to Video,” “Frames to Video,” and “Extend,” bringing voice and sound to scenes that were previously visual-only.
In Flow’s editing environment, creators can now:
- Insert additional elements into any scene with seamless handling of lighting and shadows.
- Remove unwanted objects or characters, with the system reconstructing surrounding backgrounds to maintain scene integrity.
The update also expands Flow’s capabilities in narrative control:
- Ingredients to Video enables creators to combine multiple reference images to craft the visual composition, reflecting style, objects, and characters with precision.
- Frames to Video lets users define a starting and ending image, with Flow generating intermediate frames and smooth transitions.
- Extend allows clips to be lengthened by continuing motion and action beyond the original endpoint.
Also Read: RealSense Raises $50M, Teams Up with NVIDIA for AI Robotics
The Veo 3.1 model is now accessible to developers via the Gemini API, to enterprise customers via Vertex AI, and via the Gemini app for direct use. Support for “Ingredients to Video,” “First and last frame,” and “Scene extension” is now live in the Gemini API; “Add object” and “Remove object” will roll out later. Scene extension is slated for launch in Vertex AI in a subsequent release.
Jess Gallegos, Senior Product Manager at Google DeepMind, commented: “We’re always listening to your feedback … We heard that you want more artistic control within Flow, with increased support for audio across all features.”
Thomas Iljic, Director of Product Management at Google Labs, added: “With more precise editing capabilities, audio across all existing features and higher-quality outputs powered by Veo 3.1, we’re opening up new possibilities for richer, more powerful video storytelling right inside Flow.”
