Who makes Gemini video understanding?

Gemini video understanding is made by Google.

Does Gemini video understanding work in the EU?

Available in the EU. No explicit evidence found on the fetched pages; however, the Gemini app and Gemini API are generally offered in EU/EEA countries, though some features can roll out later there. Last verified 14 Nov 2026.

How much does Gemini video understanding cost?

Freemium. Gemini API has free and paid tiers per the cookbook; the Gemini app has its own plans. Exact prices not in provided material. Licence: Proprietary (cookbook examples: Apache-2.0).

Where was Gemini video understanding shown?

In 1 creator video by Matt Wolfe. Most recently by Matt Wolfe on 9 Sept 2026, at 1:55 in "I Built A Tool To Detect AI Video (You Can Have It)".

Gemini video understanding

by Google

Try it Website Docs GitHub

Gemini API capability for analyzing video input, including visual and audio content, with text output.

Available in EULanguage Models Video GenerationModel

Updated 14 Nov 2026EU status checked 14 Nov 2026Last shown 9 Sept 2026

Gemini video understanding at a glance

What is Gemini video understanding?: Gemini video understanding is a model from Google. Gemini API capability for analyzing video input, including visual and audio content, with text output.
Who makes Gemini video understanding?: Gemini video understanding is made by Google.
Does Gemini video understanding work in the EU?: Available in the EU. No explicit evidence found on the fetched pages; however, the Gemini app and Gemini API are generally offered in EU/EEA countries, though some features can roll out later there. Last verified 14 Nov 2026.
How much does Gemini video understanding cost?: Freemium. Gemini API has free and paid tiers per the cookbook; the Gemini app has its own plans. Exact prices not in provided material. Licence: Proprietary (cookbook examples: Apache-2.0).
Where was Gemini video understanding shown?: In 1 creator video by Matt Wolfe. Most recently by Matt Wolfe on 9 Sept 2026, at 1:55 in “I Built A Tool To Detect AI Video (You Can Have It)”.

Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.

About

Gemini video understanding is the ability of Google's multimodal Gemini models to accept video as input and answer questions about it, summarize it, describe scenes, or extract information. It is documented as part of the Gemini API and is also reachable through the Gemini app and Google AI tooling. Output is text.

The Gemini API cookbook (Apache-2.0 licensed examples) lists current models such as Gemini 4 Hydrogen, Carbon and Argon, which have a 1M-token context window and configurable thinking levels. Long context is what makes analysis of longer videos practical. Developers call the API with uploaded or referenced video files and prompts.

Limits: a YouTube creator (Matt Wolfe) reported that in his own test, Gemini repeatedly found no clear AI indicators even in obvious AI-generated videos, so it should not be relied on as an AI-video detector. The models are proprietary and accessed through Google's hosted services. Specific video length, frame sampling and token cost details are in the linked docs and were not in the provided material.

Key features

Video input analysis via the Gemini API
Summarization, Q&A and scene description over video
Large context window (1M tokens on Gemini 4 models)
Configurable thinking levels on current models
Cookbook examples under Apache-2.0

1 moment

As seen on YouTube

Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.

Matt Wolfe9 Sept 2026

Live demo

I Built A Tool To Detect AI Video (You Can Have It)

Google's Gemini models, with a new video understanding capability rolled out in September, were used in the detector to analyze videos. In testing Gemini repeatedly found no clear AI indicators, even on obvious AI videos.

1:55Open on YouTube