Google and DeepMind added agentic video understanding to the latest Gemini models. Instead of statically sampling one frame per second, Gemini reasons over transcript, audio, and frames and dynamically adjusts frame rate, cutting tokens by up to 88% on long-form video. The feature is live via the Gemini API for 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite.

Key Takeaways

  • ✓Replaces static 1-fps scanning with dynamic frame-rate control over transcript, audio, and visual frames
  • ✓Biggest gains on long-form video, using up to 88% fewer tokens with higher accuracy
  • ✓Rolling out now in Gemini API / AI Studio; coming to the Gemini app and YouTube’s Ask YouTube
⚡

Key Decision Metrics at a Glance

TL;DR Verdict
🏆 Benchmark RankRank
Biggest gains on long-form video, using up to 88% fewer tokens with higher accuracy
Full Leaderboard→
💰 Cost & PlansCompare
in: $0.75 / out: $3.75
Compare Plans↗
🛠️ Ecosystem & IDERules
AI · LLM · Gemini
Setup IDE Rules↗
🧭

Turn your technical choice into a development budget

Compare 40 dev plans & simulate token costs vs $20/mo subscriptions

⚔️ Compare Model→💰 Compare plans & limits↗
Action HubReady to adopt this in production?

Benchmark side-by-side against alternatives, or calculate monthly token cost vs subscription break-even.

⚔️ Launch VS Compare→💳 View Plans↗
Up NextGoogle DeepMind Adds Agentic Video Understanding to Gemini, Cutting Tokens up to 88%
Read Now→