Google Gemini Gains Smarter Video Processing with Agentic Understanding
Official posts describe dynamic selection of frames, audio, and transcripts for long video analysis.
TLDR
Google and DeepMind accounts announced agentic video understanding for the latest Gemini models. Instead of fixed frame rates, the models reason over transcripts, audio, and selected frames to handle long videos. The posts state this approach improves accuracy while cutting token use and costs. The feature is available now through the Gemini API in AI Studio and the Enterprise Agent Platform. It is scheduled to reach the Gemini app and YouTube's Ask YouTube feature later.
Combined views
1.1M
11 Sources, first seen ago
Sources
GO 
Google@GoogleWe’re introducing a new capability to our latest Gemini models: agentic video understanding. This allows developers to process long-form video content with more accuracy, while using up to 88% less tokens. See how it works 🧵
- likes: 5.5K
- replies: 193
- bookmarks: 2.4K
- reposts: 462
PS Philipp Schmid@_philschmidGemini video understanding is now agentic. Gemini can now iteratively navigate video timelines, decide watch what, pick frame rates, or chooses whether it needs speech transcripts, audio, or visual frames to answer your prompt. Result: Long videos get up to 88% fewer tokens and…
- likes: 266
- replies: 11
- bookmarks: 126
- reposts: 13
LK Logan Kilpatrick@OfficialLoganKIntroducing Agentic Video in the Gemini API, a new way to process long videos which reduces token consumption by up to 88% while also increasing quality. This can be controlled easily in the API on a per video basis, available with our newest models like 3.7 Flash!
- likes: 2.3K
- replies: 120
- bookmarks: 558
- reposts: 119
GA Ghost AI@Ghostaisystems100% i own video production company and we have been trying to build something like this for about 6 months and today we used Fable 5.1 attached with the agentic understand and we dumped 400 GB 3 cameras multiple mics of raw footage and it gave us completed rough draft in 4…
SP srinath pulaverthi@Srinath_9991@Google @GoogleDeepMind Yo bro this is what I'm talking about, GOAT of multimodal input. This is what makes Gemini stand apart
RV Rahul Vyas@Rahul_oog@_philschmid Awesome!