What it is
Gemini is Google's general-purpose AI assistant, used through a web app, mobile apps and, since September 2026, a Windows desktop app that opens with Alt + Space. Google's overview page describes it as an interface to a multimodal large language model, so it can take text, images and audio as input.
Who it suits
It fits people who already live in a Google account and want one assistant for drafting, summarising documents, debugging code, generating images and researching topics. Google's Gemini overview says responses can draw on external sources such as Google Search and extensions.
Models and plans
The free tier gives access to Gemini 3.6 Flash and limited Gemini 3.1 Pro use, according to Google's subscriptions page. Paid plans raise usage limits and bundle Google storage and other Google products. The page listed UK prices in pounds when we checked.
Where it falls short
Google's own overview lists limits: accuracy problems on complex topics, bias inherited from training data, and vulnerability to adversarial prompting. Some premium features are restricted to users aged 18 and over, and Ultra, the priciest plan, offers the highest usage limits (up to 20x Pro).
