Google's Veo 3 and Gemini Omni Flash on ALStudio
Google has two video models that matter for brands. Veo 3, launched in May 2025, made video with built-in sound mainstream: dialogue, sound effects and ambience generated with the picture. Gemini Omni Flash is Google's newer model for fast video creation and editing through conversation.
Veo 3: key features
- Native audio: dialogue, sound effects and background sound, synced to the video.
- Realism and physics with strong prompt following.
- Text-to-video and image-to-video.
Gemini Omni Flash: key features
- Any input: combine text, images, video and audio to create video.
- Edit by chatting: refine a clip in several turns ("make it night", "move the camera closer").
- Extend, upscale and interpolate clips.
- Up to 4K output per Google's docs; newer 1.2 version supports 15-second clips.
- All outputs carry Google's invisible SynthID watermark.
Best for: realistic ads with dialogue, product scenes with natural sound, and quick versions of one idea.
How to use it on ALStudio: choose Veo 3 or Gemini Omni Flash, add references from Brand DNA, write the scene and the lines, and generate.
FAQ
What is Veo 3?
Google DeepMind's video model that generates video with native audio.
What is Gemini Omni Flash?
Google's fast multimodal model for generating and editing video through conversation.
Does Veo 3 make sound?
Yes, including dialogue and sound effects.

