Key takeaways
- Canvas is Gemini's side-by-side workspace for drafting and editing documents and code, with live previews of web apps.
- Notebooks in Google's ecosystem usually means NotebookLM, where you upload sources and ask questions answered only from them.
- Use the Pro tier or a thinking mode for hard coding and analysis, and Flash for fast everyday work.
- Gemini launched as Google's model family in December 2023, and the Bard chatbot was renamed Gemini in February 2024.
- In Gmail and Slides, Gemini drafts emails and slide content from short prompts; you still edit for accuracy and tone.
Gemini Canvas is a workspace inside the Gemini app where you write and edit documents or code alongside Gemini, with the draft in its own panel and live previews for web code. Notebooks, in Google's ecosystem, usually means NotebookLM, which answers questions only from the sources you upload. This page explains how both work, which Gemini model to pick for general work and for coding, when Gemini launched worldwide and in India, and how Gemini helps draft emails in Gmail and slides in Google Slides.
What does Gemini Canvas do?
Canvas gives you a dedicated editing space next to the chat. Gemini writes a document or code there, and you refine it together: highlight a section to change, adjust tone or length, or run and preview code such as a web page.
It is useful for longer pieces you will iterate on: an essay outline, a project report, a cover letter, a Python script or a small HTML app. For code, Canvas can show a live preview of web projects, so you can see a landing page or a simple game working and ask for changes. Canvas can also turn content into other formats, such as a web page or quiz, and export documents to Google Docs. It replaces the clumsy pattern of copying code from the chat into a file and back.
How does Gemini Canvas work?
Select Canvas in the Gemini prompt bar, describe what you want, and Gemini opens the draft in a side panel. You then chat to revise it, or edit the draft directly, and Gemini updates the canvas rather than writing a new answer each time.
Useful moves inside Canvas:
- Highlight a paragraph and ask for it to be simpler, shorter or more formal.
- Ask for a specific change: "add error handling to the fetch call".
- Preview web code and describe what looks wrong.
- Export a document to Google Docs to keep editing there.
Keep the scope small, one document or one small app per canvas, and save versions you like before asking for big changes.
What are Gemini notebooks?
In Google's AI tools, notebooks usually refer to NotebookLM: a research assistant where you create a notebook, add sources such as PDFs, Google Docs, websites or YouTube videos, and ask questions that are answered only from those sources, with citations.
Google has been connecting NotebookLM with the Gemini app, so you can work with notebook material inside Gemini as well. For students, NotebookLM is one of the best free study tools: add your lecture notes, textbook chapters and past papers, then ask for summaries, explanations, study guides or quizzes. Because answers come from your sources and link to the exact passages, they are easier to trust and check than general chatbot answers. It also creates audio overviews, a podcast-style discussion of your material.
How do Gemini notebooks work?
You create a notebook in NotebookLM, upload or link sources, and ask questions. NotebookLM retrieves the relevant passages from your sources and has Gemini answer from them, citing each claim back to the source text.
This is retrieval-augmented generation scoped to your own material, the same principle behind the document chat I built in the OCR & Speech Workspace, where answers cite page numbers. Tips: keep one notebook per subject or project, add only sources you trust, and ask specific questions such as "compare the two definitions of normalisation in these notes". Click the citations to verify. Generated study guides and quizzes are good for revision, but check them against the sources before an exam.
Which Gemini model is best?
For everyday tasks, Flash is best: fast and capable. For hard problems such as complex coding, maths and analysis, the Pro tier or a thinking mode is best. The newest generation of each tier is usually better than the previous one.
The Gemini app chooses a model automatically unless you pick one, and the options shown depend on your plan. Developers choose explicitly in the Gemini API: Flash-Lite for the cheapest high-volume tasks, Flash as the default for most products, and Pro for the hardest requests. A good production pattern is to start with Flash, measure where it fails, and route only those cases to Pro. Google's models page lists the current versions, which change several times a year.
Which Gemini model is best for coding?
Use the most capable Pro model available to you, ideally with thinking enabled, for debugging, architecture and multi-file changes. Flash is fine for quick snippets, explanations and simple scripts.
For real projects, the model matters less than the tool. Google's coding tools, such as Gemini CLI, Gemini Code Assist in your editor and Antigravity, can read your whole project, run commands and tests, and propose edits, which makes their output far more reliable than code from a chat window. Whatever you use, give precise tasks, include error messages, keep a project context file with build and test commands, and review every change. Compare against other coding tools on one real task before settling.
When was Gemini AI launched?
Google announced the Gemini model family in December 2023. Its chatbot Bard, launched in early 2023, was renamed Gemini in February 2024, when the Gemini app also arrived on Android and iPhone.
Also asked as: when gemini ai was launched · when gemini launched in india
Since then Google has released several model generations with improvements in reasoning, coding, long context and multimodal understanding, added Gemini to Gmail, Docs, Sheets, Slides, Android and Chrome, and launched developer tools including AI Studio, the Gemini API, Gemini CLI and Antigravity. The technical report on the first generation describes models trained to handle text, images, audio and video together, a design choice that shapes how Gemini works with media today.
When did Gemini launch in India?
Google's AI chatbot was available in India from 2023, under the Bard name, and the Gemini app launched in India in 2024 with support for English and several Indian languages.
Language support has expanded since, and Gemini works in Hindi, Bengali, Tamil, Telugu, Marathi, Gujarati, Kannada, Malayalam and Urdu, among others, which makes it useful for students and users more comfortable in their own language. Google has also run offers in India, including free plans for eligible students and a partnership with Jio. On Android phones in India, Gemini increasingly replaces Google Assistant as the default assistant. Check Google's Gemini help pages for the current list of languages and features.
How does Gemini help users draft an email?
In Gmail, Gemini drafts an email from a short prompt with "Help me write", suggests replies in context, and can refine a draft to be more formal, shorter or more detailed. It can also summarise a long thread before you reply.
Also asked as: how does gemini assist users in drafting an email
Write prompts that include the recipient, purpose and tone, for example "email my professor asking for a two-day extension for the project, polite and brief, mention I was unwell". Review the draft for facts: Gemini may add details or commitments you did not intend. For important emails, read it aloud once; if it does not sound like you, edit it. In Google Workspace accounts, administrators control which Gemini features are available.
How does Gemini AI help users in slide creation?
In Google Slides, Gemini can generate slide content and layouts from a prompt, create images for slides, and help rewrite text. It speeds up the first draft, while you still shape the story and check every fact.
A good workflow: write a one-paragraph brief for the deck, its audience and key message; ask Gemini for an outline of eight to ten slides; generate slides one at a time and edit them; add your own data, screenshots and examples; and cut words, since slides work best with very little text. For college presentations, prepare speaker notes in your own words, because evaluators ask questions about what you present. Gemini-generated images should be labelled when they could be mistaken for real photos.
Sources
GenAI engineer and AI product builder with 2+ years shipping production-grade AI systems: RAG pipelines, fine-tuned LLMs, hybrid retrieval and multi-modal apps across vision, speech and OCR, architected end to end from ingestion to deployment. Leads engineering for SCULT INDIA's 14-member team, founded the 500+ member TechVerse Enclave community and has mentored 200+ students. Three hackathon first prizes: Changethon 2025 (IIT Roorkee), Product Genesis at Vividhotsava 2025 (CSJMU) and BYTEBATTLE (MeetKats).
Open to GenAI roles, hackathon judging, mentorship sessions and guest talks at colleges. For campus sessions the inviting institution covers the visit end to end: travel (a First AC train ticket for journeys up to five hours, a flight for anything longer), accommodation, meals, and pickup and drop; the honorarium and workshop fees are negotiable on WhatsApp or on the call, and requests are welcome from institutions that can cover the visit this way. Email pranjulrathour41@gmail.com.
