Google Adds Agentic Video Understanding to Gemini Models
Google has introduced a new agentic video understanding feature in its Gemini models, allowing for more efficient processing of long videos. This technology reduces token costs and improves processing accuracy.
Google Adds Agentic Video Understanding to Gemini Models
Google has introduced a new feature in its latest Gemini models — agentic video understanding. Now developers can process long videos more accurately while consuming up to 88% fewer tokens.
Agentic video understanding allows Gemini to process videos dynamically — the model decides which segments to explore in more detail, reviewing frames, audio, and transcripts, and adjusting processing speed using built-in tools.
According to Google, when working with long videos, the new approach provides:
- Up to 88% fewer tokens
- Up to 66% lower costs
- Up to 7% higher accuracy
This feature is already available through the Gemini API and in Google AI Studio.
Why it matters
AnalysisThis new capability significantly enhances video processing efficiency, marking an important step in AI technology development. The reduction in token costs and improvement in accuracy opens new opportunities for developers.
Discuss in community
Share your questions and insights with developers
Related in Ecosystem
Practical guides, core terminology, and agent skills
Google AI Studio: Free Gemini API Keys
Official Google sandbox: get free Gemini 2.0/1.5 API keys without a credit card, setup system prompts, and test the 2M token context window. Limits & guide 2026.
Google releases Gemini Omni 1.2 Flash: videos with sound for up to 15 seconds
Google's updated Gemini Omni 1.2 Flash model generates and edits short videos with audio for up to 15 seconds, with 720p API pricing about 30% lower than 1.1 Flash.
Gemini Flash & Pro (Google Gemini)
A family of multimodal models from Google DeepMind that combines a record context window (up to 2 million tokens), extreme generation speed (over 150 tokens/sec), and native perception of video and audio.
Claude Sonnet 4.6 vs GPT-5.5 vs Gemini 3.1 Pro: 2026 model comparison
Every few weeks, a new model comes out, and every time the same questions arise: should I switch? what changed what to pay for?