Google DeepMind launches agentic video understanding for Gemini models
Google DeepMind announced the launch of agentic video understanding for its Gemini models, including Gemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite. The feature is available via the Gemini API in Google AI Studio and the Gemini Enterprise Agent Platform. It uses native video tools to dynamically search, scan, and inspect video segments, reducing token consumption by up to 88%, costs by up to 66%, and improving accuracy by up to 7% on standard benchmarks. The feature is particularly beneficial for long-form video analysis. It will also roll out to the Gemini app and power YouTube's 'Ask YouTube' feature in the coming months.
Comments 0
Discuss this event in persistent threads. Live chat remains separate.
What we know
The feature is available via the Gemini API in Google AI Studio and the Gemini Enterprise Agent Platform.
▤ 1 sources›
Agentic video understanding reduces token consumption by up to 88% and costs by up to 66%.
▤ 1 sources›
The feature will roll out to the Gemini app and power YouTube's 'Ask YouTube' feature.
▤ 1 sources›
Google DeepMind launched agentic video understanding for Gemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite.
▤ 1 sources›
Accuracy improves by up to 7% on standard video analysis benchmarks.
▤ 1 sources›
No additional feature fee; uses standard Gemini API token pricing.
▤ 1 sources›
Google DeepMind announced the launch of agentic video understanding for Gemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite, claiming up to 88% token reduction, 66% cost reduction, and 7% accuracy improvement.
Verified · 1 sources
No comments yet. Start the conversation.