Alphabet's Google launched agentic video understanding across its latest Gemini models. This aims to deliver higher accuracy at lower costs.
The new feature builds on agentic vision. It combines Gemini's native video processing tools with code execution. This enhances performance.
This enables advanced capabilities. These include sub-second moment retrieval, precise object counting, and more accurate anomaly detection within video content.
The upgrade aims to improve how AI models interact with and analyze video. This makes video analysis an active investigation.
The feature is now available for developers. They can access it through the Gemini API in Google AI Studio. It is also available via the Gemini Enterprise Agent Platform.