🔙 Go back to more models
AI Generator
Veo 3
High-fidelity clips with synchronized audio
Veo 3.1 is Google’s current video generation model, replacing the now-discontinued Veo 2. Per Google DeepMind’s own model page, it generates 8-second clips by default with native audio built directly into generation — dialogue, sound effects, and ambient sound synced to the video, not added afterward — and supports resolutions up to 4K.
Beyond straightforward text-to-video, Veo 3.1 supports image-to-video, character consistency across shots, scene extension, object insertion and removal, camera and motion controls, and style-reference matching — more directorial control than a single-shot prompt-to-clip model.
- Short-form marketing and social clips with built-in audio, no separate voiceover or sound-effects pass
- Multi-shot scenes that need a consistent character across cuts
- Extending or continuing an existing generated clip
- Removing or replacing objects within a generated scene
- Cinematic pre-visualization with camera and motion control
