Gemini Adds Agent-Based Video Analysis, Cuts Token Usage by 88%

The Decoder · rss · 2026-09-02

Google introduced agent-based video analysis to Gemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite. Instead of fixed-rate frame scanning, the model autonomously decides which segments to inspect and at what resolution. This cuts token usage by up to 88% while improving accuracy, particularly for long-form footage.

Original post →

More from Infra

Infra channel →