Google launches agentic video understanding in Gemini Flash models

Google DeepMind introduced agentic video understanding across its latest Gemini Flash models. The system uses native video tools so the model dynamically selects frames, audio segments, and transcripts instead of fixed-FPS ingestion. Benchmarks show up to 88% lower token consumption, up to 66% lower analysis costs, and up to 7% higher accuracy, with the largest gains on long-form video.