Gemini Adds Agentic Video Understanding, Cuts Token Usage by 88%

GoogleDeepMind · x · 2026-09-02

Google DeepMind is introducing agentic video understanding to the latest Gemini models. By dynamically adjusting frame rates and combining transcript, audio, and visual analysis, it improves accuracy while using up to 88% fewer tokens.

Related event: Google DeepMind rolls out Agentic Video Understanding for Gemini, cutting token use up to 88%(12 posts)→

Original post →

More from Models

Models channel →