Google DeepMind announces multimodal generative model “Gemini Omni,” enabling video generation and editing through natural-language dialogue and reasoning

Google DeepMind unveiled Gemini Omni, a new multimodal model family for generating and editing video from diverse inputs.

The model supports reference-based editing, natural language control, and AI avatars that can use a user's voice.

Google also said generated videos will include SynthID watermarking for transparency.

The update aims to make consistent, text-driven video creation more practical for creators and everyday users.