Gemini, by Google

What is Gemini?

Gemini is Google DeepMind's AI assistant — the one that shows up as an "AI Overview" at the top of Google Search results, lives inside Gmail and Google Docs, and has its own standalone chat interface at gemini.google.com. Unlike most AI tools that were purely text-first, Gemini was built multimodal from day one: it understands images, PDFs, audio, and video alongside text.

What makes Gemini distinctive is not raw capability — it's integration. Every Google product you already use is quietly getting Gemini woven in. If you spend your days in Gmail, Docs, Sheets, and Drive, Gemini is the AI assistant that meets you where you already are.

Models at a glance

The context window is Gemini's standout spec — 1 million tokens means you can upload entire books, codebases, or hours of video transcripts and ask questions about all of it.

Free vs Advanced

The magic moment

Open Gmail, find a long email thread you haven't had time to read, and click the Summarize this email button that Gemini adds to every thread. It reads the entire conversation and tells you in two sentences what is being asked of you. If you've been using email for years and suddenly don't have to read the thread — just the summary — that's when Gemini stops feeling like a feature and starts feeling indispensable.

Step-by-step setup

1. Go to gemini.google.com — just needs a Google account
2. Sign in with your existing Gmail or Google account — no new account needed
3. Try your first prompt: paste a document URL, upload an image, or just ask a question
4. Inside Gmail: open any email thread and look for the Gemini sparkle icon in the toolbar
5. Inside Google Docs: open a document, click Help in the menu, then Ask Gemini
6. Optional: subscribe to Google One AI Premium at $19.99/month for Gemini Advanced and full Workspace integration

No app downloads required — it runs entirely in your browser.

Key features

Google Workspace integration is Gemini's killer feature. In Gmail it can draft replies, summarize threads, and extract action items. In Docs it can help write, rewrite, and proofread. In Sheets it suggests formulas and helps analyze data. None of these require switching tabs or copy-pasting — Gemini reads the content you're already looking at.

Multimodal understanding means you can upload a photo of a whiteboard, a screenshot of a chart, or a picture of a receipt and ask questions about it in natural language. Gemini reads them. ChatGPT can too, but Gemini's image understanding is particularly strong for documents and structured visual information.

Real-time web access is on by default — even on the free tier. When you ask a question, Gemini searches the web and shows you sources alongside its answer. This is a significant advantage over Claude and the free tier of ChatGPT.

NotebookLM connection — Gemini and NotebookLM are separate tools but work well together. Use Gemini for day-to-day questions and Google Workspace tasks; use NotebookLM when you want to deeply study a specific set of documents you upload.

How to get better results

Be specific about format. "Summarise this in three bullet points for a non-technical audience" gets a much more useful result than "summarise this."

I have a meeting with a client tomorrow. Read this email thread and:
1. Tell me the three things they are most concerned about
2. Suggest two questions I should ask
3. Draft a one-paragraph agenda for the meeting

Use the Google Drive integration (Advanced) to work with files you already have stored. "Analyse the Q3 sales report in my Drive and tell me which region underperformed" beats copy-pasting.

Use by profession

Teachers — generate lesson plans, differentiate material for different ability levels, create quiz questions from any topic, and summarize curriculum documents uploaded as PDFs.

Personal assistants and admins — draft and tidy emails, create meeting agendas from bullet notes, extract action items from long email chains, and draft status updates.

Marketers — use Gemini inside Google Slides to refine presentations, summarize competitor research, and draft social copy that matches your existing brand voice from a Google Doc.

Researchers — upload academic PDFs and ask questions across multiple papers simultaneously. The 1M token window makes Gemini one of the few tools that can actually fit long research documents into a single context.

Business owners — analyze spreadsheets and sales data in Google Sheets, draft contracts or proposals in Docs, and summarize long email threads from suppliers or partners.

Gemini vs the alternatives

Pick Gemini if you live inside Google Workspace and want an AI that works with your existing files and emails without any copy-pasting. Pick ChatGPT for the widest ecosystem and coding help. Pick Claude for long, nuanced writing tasks. Pick Perplexity when you need heavily sourced research.