# Google DeepMind — Agentic video understanding

- Company: Google DeepMind (deepmind.google)
- Announced: 2026-09-01
- Category: capability-change
- Coverage: not counted
- Announcement: yes
- Group: announcements
- Source: https://ai.google.dev/gemini-api/docs/changelog#09-01-2026
- Record: https://forck.live/items/24005-agentic-video-understanding
- Subject: Gemini API
- Models affected: Gemini 3.7 Flash, Gemini 3.6 Flash, Gemini 3.5 Flash-Lite

Google DeepMind added agentic video understanding to Gemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite across the Interactions and GenerateContent APIs. The model dynamically navigates video timelines, requesting transcripts, frames, or audio tracks on demand instead of processing every frame. This approach uses up to 88% fewer tokens for long-form content compared to static processing, according to Google DeepMind. The feature is available now through the Gemini API, with a guide to get started.

## Evidence

Verbatim from https://ai.google.dev/gemini-api/docs/changelog#09-01-2026:

> The model dynamically navigates video timelines, requesting transcripts, frames, or audio tracks on demand. This approach uses up to 88% fewer tokens for long-form content compared to static processing.

---

Record: https://forck.live/items/24005-agentic-video-understanding
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md