Google Gemini: Multimodal and Integrated
You understand Gemini's strengths and Google Workspace integration.
Why this lesson matters
Google Gemini is the third big player in AI, and it has one unique advantage: deep integration with Google Workspace. If your organisation runs on Gmail, Google Docs, Drive, Sheets and Calendar, Gemini is an AI that already knows your work context.
Gemini is also multimodal (text, image, audio, video) and, at 1 million tokens, has the largest context window.
Google Workspace integration
Gemini is built into the tools you use every day:
| Google tool | What Gemini does |
|---|---|
| Gmail | Summarise emails, draft replies, adjust tone |
| Docs | Generate text, rewrite, summarise, expand |
| Sheets | Generate formulas, analyse data, suggest charts |
| Slides | Generate presentations from a prompt or document |
| Drive | Search and summarise files |
| Meet | Summarise meetings, extract action items |
Example in Google Sheets:
Select a column of sales data and ask Gemini:
"Create a formula that calculates the moving average
over the last 3 months, and flag months that deviate more than
20% from the average."The Workspace integration works with Gemini for Google Workspace (formerly Duet AI). This is a paid add-on for Google Workspace accounts. The standalone Gemini app (gemini.google.com) is partly available for free.
Multimodal: text, image, audio and video
Where Claude and ChatGPT mainly work with text and images, Gemini also understands audio and video natively.
Analysing video:
Upload a product video and ask:
"Analyse this demo video. Give me:
1. A summary of every feature shown
2. Timestamps of the key moments
3. Suggestions to improve the presentation"Keep reading
Leave your name and email address and you can read the rest
You get the whole lesson right away, and every other lesson stays open after that. No password, no confirmation email.
Processing audio:
Upload a recording of a meeting:
"Write minutes for this meeting. Structure:
- Participants (if identifiable)
- Topics discussed
- Decisions taken
- Action items with owner"The multimodal capabilities make Gemini ideal for teams that work a lot with video and audio: marketing, sales, training and communications. Upload a webinar and get a complete summary within minutes.
NotebookLM: your personal research assistant
NotebookLM is a separate Google AI product you should know about. It's a research tool where you upload documents and get an AI assistant that answers only from your sources.
What makes NotebookLM unique:
- Upload up to 50 sources (PDFs, websites, YouTube videos, Google Docs)
- The AI answers only from your sources, no hallucinations
- Every claim has a source reference you can verify
- Generate summaries, FAQs and audio conversations automatically
Use case:
1. Upload 10 research reports on a topic
2. Ask: "What are the main contradictions between these sources?"
3. NotebookLM answers with quotes and references to specific documents
4. Click the reference to read the original passageNotebookLM is a different product from Gemini itself. You find it at notebooklm.google.com. It's free to use and doesn't need to be linked to a Workspace account.
A million tokens: what a large context window gives you
The context window is the amount of text a model can read in one go. The largest models from Anthropic, OpenAI and Google are all around a million tokens by now, roughly 2,500 pages. Who leads on that point changes every quarter, so check the current limit on the provider's page rather than here.
What can you do with a million tokens?
- Analyse a complete codebase in one go
- Upload an entire book and ask questions about it
- Search through hours of meeting minutes at once
- Compare dozens of contracts in a single session
When to choose Gemini?
| Situation | Gemini is the best choice if... |
|---|---|
| Your organisation runs on Google Workspace | The integration saves an enormous amount of time |
| You work a lot with video and audio | Native multimodal processing |
| You need to analyse huge amounts of text | 1M context window |
| You want AI inside your existing workflow | Gemini is already in your tools |
| You want to do source-based research | NotebookLM with source references |
When Gemini is not the best choice: complex reasoning (Claude is stronger), generating images (DALL-E is better), debugging code (Claude Code wins), or if you don't work with Google Workspace.
Practice exercise
- 1Open gemini.google.com and ask the same question you asked Claude earlier, compare the answers
- 2If you have Google Workspace: open Gmail and try Gemini's summarise function on a long email
- 3Go to notebooklm.google.com, upload 3 documents and ask cross-document questions
- 4Upload a video to Gemini and ask for a summary with timestamps
What is Gemini's main advantage over Claude and ChatGPT?
What makes NotebookLM unique compared to standard AI chatbots?
Want to keep your progress, get the practice files and sign up for the free evening? Create a free account. All you do is click a link in your email.
Create a free account