Google Adds Agentic Video Understanding to Selected Gemini Models

On September 1, 2026, Google announced agentic video understanding for the Gemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite models. It is designed to analyze uploaded videos and YouTube videos through the Gemini API in Google AI Studio and the Gemini Enterprise Agent Platform.

The new mode is intended to change how the model searches for information when working with video. Instead of processing content at a preset frame rate, it is supposed to dynamically find and check relevant segments across visual frames, the audio track, and transcripts.

What agentic video understanding offers

With conventional static video processing, the developer sets what portion of the content or how many frames the model will evaluate. With long recordings, this increases the number of processed tokens and costs, while an important detail may remain outside the selected sample.

Google describes the new approach as an internal agentic process within the model. According to the specified question, Gemini is supposed to first search for potentially relevant sections and then verify them across the image, audio, and text transcript. The selection of segments thus moves from manual implementation in the developer’s application into the model itself.

The feature is activated by setting processing to “agentic” in the API configuration. Google says it will not charge a separate premium for this mode: the standard Gemini API token prices are to be used.

Available models and deployment

Google announced the agentic video analysis mode for three models in the Flash family:

• Gemini 3.7 Flash,
• Gemini 3.6 Flash,
• Gemini 3.5 Flash-Lite.

Developers can use it with videos uploaded to the system and with YouTube videos. According to the announcement, it is available through the Gemini API in Google AI Studio and the Gemini Enterprise Agent Platform. The announcement does not specify detailed feature limits or complete processing parameters.

Benchmarks currently come from Google

In its presentation, Google reports up to an 88% reduction in token usage and up to a 66% reduction in costs, along with an accuracy increase of up to 7%. These are, however, the manufacturer’s benchmark figures. No independent verification of the methodology or results was provided with the announcement, so it is unclear how these numbers will translate to different types of videos and practical deployments.

The feature’s significance lies mainly in the fact that video analysis in multimodal models is often token-intensive. If the stated results are also confirmed outside internal benchmarks, developers could search for specific events, audio information, or visual details in longer videos at a lower cost.

What comes next

Google also announced plans to expand the feature to the Gemini app and the “Ask YouTube” service. The company has not yet confirmed the timing or scope of this availability beyond its own announcement.

As the feature is deployed further, independent comparisons of accuracy, latency, and real-world costs against static video processing will be important to watch. Questions also remain about API limits, documentation, and privacy protection when processing video content.

Verified and updated: 09/01/2026 22:31