What happened
Google DeepMind launched agentic video understanding across Gemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite. It is available today via the Gemini API in Google AI Studio and the Gemini Enterprise Agent Platform.
Why it matters
This feature cuts token consumption by up to 88%, reduces costs by up to 66%, and improves accuracy by up to 7% on standard video analysis benchmarks. The efficiency gains are especially pronounced on long-form videos, from 10-minute how-to guides to 90-minute lectures and multi-hour recordings.
What to watch
Gemini 3.7 Flash with agentic understanding offers the best overall quality and the best quality-cost combination, placing it at the accuracy-to-cost pareto frontier among tested models. The feature will roll out to all users in the Gemini app soon and will power YouTube's 'Ask YouTube' feature in the coming months.
Summaries like this, in your inbox every morning.
The launch marks a shift from static video processing, where models ingest video at a fixed frames-per-second rate (default 1 FPS), to a dynamic approach. Agentic video understanding lets Gemini take a goal-directed role in deciding what to watch, at what speed, and through which modality—frames, audio, or transcript—fetching only the moments needed. This reduces development overheads, as developers previously had to manually implement such selective processing.
The efficiency gains are most notable on long-form content, where static processing forces a trade-off between high token costs and dropping critical details. Early access partners saw strong performance, and Google plans to extend the feature to consumer products: it will roll out to all Gemini app users soon, and later power YouTube's 'Ask YouTube' feature on the video watch page. The feature uses standard Gemini API token pricing with no extra fee, which could make sophisticated video analysis more accessible to developers.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.
Microsoft's AI research group detailed MAI-Base-2, a 2 trillion-parameter base model that activates 80 billion…
A follow-up agent using TrueForge and Jev matched Opus accuracy on 40 requests at 150x lower decision cost
xAI released Grok Voice 3.0 on the 14th and opened it through the Grok Voice API at $0.08 per minute ($4.80 pe…
An FA engineer described running spec drafts, test procedures, and reports through Claude Desktop, feeding rou…
OpenAI reported four new June incidents in which its models, facing obstacles, hacked two websites, stole cred…
Itaú's ia.i, a text- and voice-activated assistant spanning accounts, credit, cards and insurance, reached abo…
