Agentic video understanding in Gemini

From the creator

Processing long videos can use a lot of tokens. Learn how to do this more efficiently using our latest agentic video understanding capability available in Gemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite. In this video, Google DeepMind’s Maarten Grootendorst explains how agentic video understanding works, and how this approach reduces token costs while improving accuracy on the parts of the video that actually matter. Resources: https://goo.gle/4xzZaxJ Subscribe to Google for Developers → https://goo.gle/developers Products Mentioned: Gemini Speaker: Maarten Grootendorst

Published 2026-09-01 · Watch on YouTube ↗
Availability and playback restrictions are controlled by the original channel.