Google Gemini Cheat Sheet: A Complete Guide

Google Gemini can answer questions, analyze files, create documents and media, and complete supported tasks across connected applications.

Its models handle writing, coding, research, and creative work, while agentic features can act on connected accounts. Available tools and usage limits depend on the subscription, device, region, and account type.

Google Gemini at a glance

What does Google Gemini do?

Some offerings are AI models that process or generate content. Others are tools for research, reusable workflows, or actions across connected applications.

Writing, coding, and research

Uploaded documents, screenshots, and other files can provide context for more detailed analysis.

• Gemini 4 Argon. Google’s advanced reasoning model handles complex coding, analysis, and multistep assignments. App access is included with Google AI Pro and Ultra; developer and business access is separate.
• Flash models. Optimized for quick responses to everyday requests such as translation, summaries, math, and programming help. The Flash family also supports developer workloads, with model options that differ from the consumer app.
• Deep Research. Investigates a topic across multiple sources and produces a referenced report.
• Gemini Notebook. Uses selected source material to produce summaries, study aids, and grounded analysis. Notebooks can sync with the app; advanced code-based analysis has separate eligibility rules.
• Canvas and file creation. Drafts and edits writing or code, generates downloadable documents, spreadsheets, and PDFs, and prepares presentations in Canvas.

Research citations still need checking, while generated code should be tested before use.

Image, video, and music creation

Creative projects may involve different prompts, reference materials, and editing approaches.

• Nano Banana. Google’s image tools generate pictures from prompts and edit objects or backgrounds in existing images.
• Gemini Omni. Generates video from text and visual references, with follow-up prompts for revisions. Omni Flash is available through eligible Gemini app and Google Flow experiences.
• Veo. Produces clips from prompts and reference images, with supported audio, camera, and scene controls. It is accessible through Google Flow and other eligible tools.
• Lyria. The Lyria 3.5 model creates instrumental tracks or songs with vocals. Prompts can specify genre, style, and track length.

Projects combining different media formats may require more than one tool.

Live and personalized assistance

Hands-on tasks can benefit from assistance that doesn’t depend on typing.

• Gemini Live and Guided Vision. Live supports spoken conversations and camera sharing, while Guided Vision adds real-time descriptions and visual guidance on compatible Android devices.
• Personal Intelligence and Daily Brief. With permission, Personal Intelligence uses information from connected Google services to personalize responses. Daily Brief assembles updates and reminders for eligible accounts.

Camera access and connected-account permissions are managed separately.

AI agents and task automation

Agents can complete supported multistep assignments in connected services, subject to each integration’s permissions.

• Gemini Spark. Handles work across authorized services, from organizing files on a Mac to researching travel options through its Chrome integration. It can begin a booking in an authorized browser session, but sensitive steps such as payments require approval. Some tasks continue in the background.
• Gemini Skills. Store instructions and reference files for recurring work, such as report formatting, and apply them across chats and tasks. Skills currently require a personal Google account, age 18 or older, and Keep Activity enabled. Google is transitioning Gems to Skills, with work and school account access to follow.
• Connected applications. Link Gemini to Google services and third-party tools including Airtable, Linear, monday.com, and Adobe.

Background tasks and browser automation have additional plan and regional restrictions.

How does Google Gemini work?

Gemini processes text, images, audio, and other supported inputs through cloud-hosted AI models. Model options differ between the consumer app, developer APIs, and enterprise products.

Compatible devices can handle selected tasks locally using Gemini Nano, a lightweight model distinct from the Nano Banana image tools. Cloud access does not guarantee support for on-device features.

A model’s context window determines how much information it can process in a request, including material from long documents or conversations.

Where is Google Gemini available?

Gemini is available through dedicated apps and other Google products:

• Web. Access the assistant at gemini.google.com in supported countries and languages.
• Android. Available on eligible phones and tablets. The Google Assistant transition involves device-specific changes.
• iPhone and iPad. The mobile app runs alongside Siri, with additional features available through Chrome on Apple devices.
• Windows and Mac. Native apps are available from Google’s Windows and Mac download pages. Open them with Alt + Space on Windows or Option + Space on Mac. Selected macOS features can use window context with permission.
• Google Workspace. Built into supported versions of Gmail, Docs, Sheets, Slides, and Meet, subject to organizational controls.
• Chrome and Search. Chrome can summarize webpages and compare tabs, while Search offers AI-powered features. Browser access has expanded across Asia-Pacific and gained Android features; selected capabilities also appear on Chromebooks.
• Smart devices. Features extend to selected Wear OS devices, Android Auto, Google TV, and compatible home products. The Gemini for Home rollout follows a separate schedule.

Features vary by region, age, subscription, and account type.

Device compatibility and requirements

Native app requirements include:

• Android: Android 9 or later and at least 2 GB of RAM. Android Go and Android Work Profiles are excluded.
• iPhone and iPad: iOS 16 or later.
• Windows: Windows 10 or later on an x64 or ARM64 device.
• Mac: Apple Silicon hardware running macOS Sequoia 15 or later.

Desktop apps need an internet connection. Some unsupported computers can still access the browser version.

Google Workspace and enterprise

Workplace uses include drafting and summarizing messages in Gemini-powered Gmail, with access governed by company policies.

Google’s agent for work uses business information and connected tools for supported workflows. Gemini Enterprise offers separate administrative and governance options, including controls over company data that agents can access.

IT teams should establish permissions, approval points, and review procedures before deploying agents across an organization.

How much does Google Gemini cost?

Prices below reflect regular US monthly rates, excluding promotions, based on Google’s subscription comparison and Ultra membership guidance. Regional prices and feature availability may differ.

Rather than a fixed daily message count, compute-based usage limits consider model choice, prompt complexity, chat length, and the feature used. Quotas can refresh every five hours up to a weekly cap; video creation, agents, and long-context work may have separate limits.

Google Workspace, Gemini Enterprise, and developer APIs have separate pricing and terms from personal plans.

How does Google Gemini handle your data?

For personal accounts, Keep Activity affects how Gemini retains and uses chats and uploads. Saved activity may be used to improve services, including training AI models. With Keep Activity off, future chats are not used to train Google’s AI models unless users submit feedback. Google still retains those chats for 72 hours for service and safety purposes, and safety-related human review can still occur. Previously reviewed chats may be retained separately for up to three years.

Personal Intelligence and connected apps request access to selected data or actions. Review those permissions before linking email, files, or calendars; turning off Keep Activity can disable some connections.

Qualifying managed Workspace accounts have different protections. Google’s Workspace privacy guidance describes restrictions on training models outside an organization’s domain without permission and options for administrators to control data access.

How do I activate Google Gemini?

1. Sign in. Open Gemini in your browser or installed app and sign in with an eligible Google account.
2. Enable the Android assistant, if needed. On supported phones, select Gemini through the device’s assistant settings.
3. Connect other services. Open Settings > Connected Apps, sometimes found under Personal Intelligence. Select the services to authorize and review their permissions.
4. For work or school accounts. Administrators may need to enable Gemini and approve integrations through organizational settings. Once access is granted, employees can sign in with their managed accounts.

What are the competitors to Google Gemini?

Gemini competes with OpenAI’s ChatGPT, Anthropic’s Claude, and Microsoft’s Copilot. All offer conversation, writing, and analysis, but their models, automation, integrations, and subscription terms differ.

Apple Intelligence works on supported Apple devices. Samsung Galaxy AI offers phone-level features, some powered by Google technology.

For individuals, app integrations and plan limits often determine which service fits best. Organizations also need to compare licensing, data protections, and approval controls for automated actions.

Want to learn more AI tips, tricks, and prompting techniques? eWeek readers get free 7-day access to The Neuron Academy, our practical learning platform designed to help professionals use AI more confidently at work. Browse all lessons →

Whether you’re using Gemini for research, writing, or everyday questions, the way you phrase your requests can affect the results. Learn how to write clearer prompts with our free six-minute How to Talk to AI course, and explore other Neuron Academy courses free for seven days. Check out all the lessons here →