Google Launches Agentic Video Understanding for Gemini Models

01 What happened

Google has introduced agentic video understanding for its Gemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite models. This feature moves away from fixed-frame-rate ingestion, using native tools to dynamically search, scan, and inspect specific visual frames, audio, and transcripts.

02 Key details

• Google claims the feature reduces token consumption by up to 88% and lowers video analysis costs by up to 66% compared to static processing.
• The company reports that the new method improves video analysis accuracy by up to 7%.
• The system replaces fixed 1 FPS ingestion with an agentic approach that targets relevant video segments directly.

03 Why it matters

This update enables developers to process long-form video more efficiently. By focusing on relevant segments rather than total frame counts, it optimizes resource usage for tasks like anomaly detection, object tracking, and sub-second retrieval.

04 Who it matters to

Software developers, data engineers and machine learning engineers.