Gemini is Google's AI assistant. You can use it at gemini.google.com, in the mobile app, and in the desktop app for macOS and Windows. You can also use it without opening it at all, because it is built into Android, Chrome, Google Search, and the Workspace apps.

Most assistants begin every conversation knowing nothing about you. Gemini begins inside the place where your mail, files, photos, and calendar already live. That one fact shapes almost everything else in this course.

Three Things Other Assistants Do Not Have

• Your own data as context: with your permission, Gemini reads from Gmail, Photos, Drive, Calendar, YouTube, and Search, so it can answer from your life instead of the public web alone;
• Creation in the same window: images, video, and music are produced inside the chat rather than in a separate product;
• Agents that work without being asked: one assembles a morning digest, another runs tasks in the background and checks with you before anything consequential.

Choosing a Model

The model picker sits in the top-left corner. Flash models are the fast default and cover most daily work. Pro models take longer and reason better on hard problems. Deep Think is reserved for the heaviest reasoning and is limited to the top plans. Version numbers move quickly, so remember the tier name rather than the number attached to it.

Plans and Limits

The free plan covers chat, uploads, image creation, and a limited amount of the heavier tools. Google AI Plus, Pro, and Ultra raise those limits and unlock video generation, the strongest models, and the newer agents. Usage is measured by the compute your requests consume rather than a fixed number of prompts per day, so one long research run costs more of your allowance than a dozen short questions.