Google Gemini
Google Gemini is Google's multimodal AI assistant, offering up to a 1M-token context window on paid plans, Deep Research, and deep integration with Google apps and services, with a Free tier now limited to the Flash-Lite model.
What is Google Gemini?
Google Gemini is Google's multimodal AI assistant, built on the Gemini model family and available through a dedicated app on web, Android, and iOS, as well as through Google AI Studio and the Gemini API. Its official positioning spans general assistance, long-document and multimodal analysis, research, and integration with Google's productivity and cloud services.
The product is organized around a tiered subscription structure. The Free tier was significantly reduced in October 2026: personal accounts without a paid Google AI subscription now only have access to the Flash-Lite model. Google AI Plus ($4.99/month) adds the Flash model and a 128K context window. Google AI Pro ($19.99/month, the successor to Gemini Advanced) unlocks Gemini 3.1 Pro, Deep Research, and a 1M-token context window. Google AI Ultra ($99.99 or $200/month) provides the highest usage multiples and storage.
A defining characteristic of Gemini is its close coupling with Google's data and infrastructure. On paid plans, it draws on Google Search, Google Workspace, and Google Cloud services, and Deep Research can synthesize web results with enterprise data through Vertex AI. At the same time, the consumer app's default data practices—training on conversations unless the user opts out—differ sharply from the paid API and Vertex AI terms, which exclude training on customer inputs and outputs.
How does Google Gemini work?
Users interact with Gemini through natural-language instructions, file uploads (documents, images, audio, video, code repositories), and, on supported plans, Deep Research. In a standard chat, Gemini processes the prompt and any attached context within the plan's context window—32K tokens on Free, 128K on AI Plus, and up to 1M tokens on AI Pro and Ultra. It returns text, code, analysis, generated media, or structured reports.
Deep Research operates differently. When given a research question, it assesses whether the prompt is research-related, generates a research plan, streams its findings as it investigates, and produces a citation-backed report alongside a 1–2 minute audio summary. It can draw on public web results and, in enterprise contexts, on indexed internal data sources. On Google AI Pro and Ultra, this agent runs on Gemini 3.1 Pro.
Gemini also integrates into Google's application surfaces. On paid plans, it can be invoked within Gmail, Google Docs, and other Workspace apps, and it connects to Google services such as Search and Maps for grounding. The output is delivered as conversational text, editable documents, code, audio summaries, or research reports, depending on the task and plan.
Use Cases
- Analyzing and summarizing long documents, entire codebases, or video/audio files up to the plan's context limit (1M tokens on AI Pro and Ultra)
- Conducting multi-step research with Deep Research and receiving citation-backed reports with audio summaries
- Drafting, editing, and working with documents inside Google Docs, Gmail, and other Workspace applications on paid plans
- Working with multimodal inputs—images, audio, video, and PDFs—alongside text prompts
- Prototyping and testing code with Gemini 3.1 Pro and the Gemini API / AI Studio (API billed separately)
- Using Gemini Flash-Lite for lightweight, high-volume tasks where the free tier's smaller context window is sufficient
Pricing
Pricing model: Freemium
| Plan | Price | Billing | Included usage | Notes |
|---|---|---|---|---|
| Free | $0 | per month | Access to Gemini Flash-Lite only for personal accounts since October 9, 2026. 32,000-token context window. Standard usage limits with a five-hour rolling session and a weekly cap. Deep Research and media generation consume the same allowance. Free-tier conversations are used to improve Google products by default, with human review of some chats. | — |
| Google AI Plus | $4.99 | per month | Access to the Flash model (not Pro). 128,000-token context window. 2x the standard usage allowance. Includes 400 GB of cloud storage. Price reduced from $7.99 in June 2026. Deep Research access is not included at this tier. | — |
| Google AI Pro | $19.99 | per month | Full access to Gemini 3.1 Pro and Deep Research. 1,000,000-token (1M) context window. 4x the standard usage allowance. Includes YouTube Premium Lite in some regions. The plan previously branded as “Gemini Advanced” is now Google AI Pro. | — |
| Google AI Ultra (base) | $99.99 | per month | 5x Pro-level usage limits. 1M-token context window. 20 TB of cloud storage. Access to Deep Think exclusive reasoning mode (requires Pro model). Price cut from $249.99 in May 2026. | — |
| Google AI Ultra (top) | $200 | per month | 20x Pro-level usage limits. 1M-token context window. For users who need the highest available capacity on Google's consumer AI stack. | — |
Common Limitations
- Free tier model restriction · As of October 9, 2026, personal Google accounts without a paid Google AI subscription can only use Gemini Flash-Lite. Gemini Flash and Gemini 3.1 Pro are no longer available on the Free tier. Users who relied on the free Pro or Flash models for long-document analysis, coding, or complex reasoning must subscribe to Google AI Plus ($4.99) or above to regain access.
- Context window gated by plan · The 1M-token context window is not available across all tiers. Free accounts are limited to 32,000 tokens (roughly 50 pages of text). Google AI Plus provides 128,000 tokens. Only Google AI Pro and Ultra unlock the full 1M-token context window. Users uploading large codebases, video files, or entire document sets should verify their plan's context limit before relying on Gemini for the task.
- Consumer training default (opt-out required) · On the Gemini consumer app (Free and paid consumer tiers), conversations are used to train Google's models and may be read by human reviewers by default for users 18 and older. Users must disable “Gemini Apps Activity” or use “Temporary Chat” mode to opt out. This default does not apply to paid Gemini API or Vertex AI usage, where inputs and outputs are contractually excluded from training.
- Deep Research is not uniformly available · Deep Research, Google's multi-step research agent that generates citation-backed reports and audio summaries, requires Google AI Pro ($19.99) or Ultra. It is not included on the Free tier or Google AI Plus ($4.99). Deep Research consumes usage allowance at a higher rate than standard chat, and heavy use can accelerate the five-hour and weekly limits.
- API and consumer subscription are separate products · A Google AI Pro or Ultra subscription enhances the Gemini app experience but does not include Gemini API or Vertex AI access. API usage is billed per token separately. Conversely, API credits are not included in consumer subscriptions. Free-tier API usage may be used for product improvement, while paid API and Vertex AI tiers are excluded from training. Users should not assume that a consumer subscription grants commercial API rights or data-protection terms.
Key Features
- Gemini 3.1 Pro model access (Google AI Pro and Ultra)
- Up to 1M-token context window (AI Pro and Ultra)
- Deep Research agent with citation-backed reports and audio summaries
- Multimodal input: text, images, audio, video, and PDFs
- Deep integration with Google Search, Maps, and Workspace apps
Google Gemini vs. ChatGPT, Claude, and DeepSeek
This comparison uses official product and pricing information available as of November 2026. It focuses on documented plan structures, context windows, model access boundaries, and privacy defaults. Gemini's value proposition is inseparable from the Google ecosystem, and its free tier has materially changed since October 2026. Plan details should be confirmed on official pricing pages before subscribing.
| Tool | Positioning | Key differences | Best fit | Pricing note |
|---|---|---|---|---|
| Google Gemini Gemini plans (official) Gemini model access changes (official support) Gemini API pricing (official) | Google's multimodal AI assistant, differentiated by long-context analysis, Deep Research, and deep integration with Google's ecosystem. | The 1M-token context window requires Google AI Pro ($19.99) or Ultra. The Free tier is limited to Flash-Lite (32K context) since October 2026. Deep Research and Gemini 3.1 Pro are gated behind AI Pro. Consumer conversations train by default, while paid API/Vertex data does not. Google AI Plus at $4.99 is the cheapest paid entry point among major assistants, but it excludes Pro and Deep Research. | Users and teams already embedded in Google's ecosystem who need long-context analysis (1M tokens) and Deep Research, and who are willing to pay for AI Pro or Ultra to get it. | Free $0 (Flash-Lite, 32K context). Google AI Plus $4.99/month. Google AI Pro $19.99/month. Google AI Ultra $99.99 or $200/month. API billed separately. |
| ChatGPT ChatGPT pricing | OpenAI's general-purpose assistant, with a broad toolset spanning writing, coding, image generation, voice, and Deep Research. | ChatGPT's entry paid tier is $8/month (Go), lower than Gemini's $4.99 Plus in absolute terms only if Go's capabilities are sufficient; the $20 Plus tier is comparable to Google AI Pro in price. ChatGPT's free tier still includes limited access to Deep Research and Codex, whereas Gemini's free tier no longer includes Pro or Deep Research. ChatGPT Plus does not advertise a 1M-token context window comparable to Gemini AI Pro. | Users who want the lowest absolute paid entry point ($8 Go) and a free tier that still includes limited Deep Research and coding agents, and who do not need a 1M-token context window. | Free available. Go $8/month. Plus $20/month. Pro tiers at $100, $200, and $500/month. API billed separately. |
| Claude Claude pricing | Anthropic's assistant for work, with Claude Code for agentic development and strong document and coding workflows. | Claude Pro at $20/month includes Claude Code and Chrome integration, which Gemini AI Pro does not replicate in the same form. Gemini AI Pro at $19.99 offers a 1M-token context window, Deep Research, and Google ecosystem integration that Claude does not provide. Both consumer products train on conversations by default, while both paid API paths exclude training. Claude does not offer a $4.99 entry tier comparable to Google AI Plus. | Developers and knowledge workers who prioritize agentic coding (Claude Code) and document workflows over Google ecosystem integration and 1M-token context. | Free $0. Pro $20/month ($17/month annually). Max $100 or $200/month. Team $25/seat/month ($20 annual). API billed separately. |
| DeepSeek DeepSeek API pricing (community-maintained comparison citing official rates) | A cost-optimized Chinese AI model family, notable for low API token prices and competitive coding performance. | DeepSeek competes on API price rather than consumer subscription breadth. Its V4 Pro and V4 Flash API models offer 1M-token input context at rates substantially lower than Gemini's paid API tiers (e.g., V4 Flash at $0.14/$0.28 per million tokens). DeepSeek does not offer a consumer assistant product with Google Workspace integration, Deep Research agents, or the same breadth of multimodal input surfaces. It is an API-first comparison rather than a consumer-assistant one. | Developers and teams optimizing API cost for high-volume text and coding workloads who do not need Google ecosystem integration or a consumer assistant interface. | API pricing only in this comparison. V4 Pro $0.435/$0.87 per million tokens. V4 Flash $0.14/$0.28 per million tokens. No consumer subscription plans comparable to Gemini AI Plus/Pro/Ultra. |
Comparison checked against official product information on 2026-11-03. Features and pricing may change.
Frequently Asked Questions
Is Google Gemini free?
Yes, but the Free tier is limited. As of October 9, 2026, personal Google accounts without a paid Google AI subscription can only use Gemini Flash-Lite. Gemini Flash and Gemini 3.1 Pro require Google AI Plus ($4.99) or above. The Free tier also has a 32,000-token context window and standard usage limits.
What is the difference between Google AI Plus, Pro, and Ultra?
Google AI Plus ($4.99) adds the Flash model, a 128K context window, 2x standard usage, and 400 GB of storage, but does not include Pro or Deep Research. Google AI Pro ($19.99) unlocks Gemini 3.1 Pro, Deep Research, a 1M-token context window, and 4x usage. Google AI Ultra starts at $99.99 for 5x Pro usage, 20 TB of storage, and Deep Think, with a $200 tier for 20x Pro usage.
Does Gemini use my conversations for training?
On the consumer Gemini app (Free and paid consumer tiers), yes by default for users 18 and older. Conversations are used to improve Google products and may be read by human reviewers. You can opt out by disabling “Gemini Apps Activity” or using “Temporary Chat” mode. This default does not apply to the paid Gemini API or Vertex AI, where inputs and outputs are excluded from training by contract.
Is Deep Research available on the free plan?
No. Deep Research requires Google AI Pro ($19.99) or Ultra. It is not included on the Free tier or Google AI Plus ($4.99). Deep Research consumes usage allowance at a higher rate than standard chat and can accelerate the five-hour and weekly limits.
Does a Google AI Pro subscription include Gemini API access?
No. A consumer Google AI subscription (Plus, Pro, or Ultra) enhances the Gemini app but does not include Gemini API or Vertex AI access. API usage is billed per token separately. Conversely, API credits are not included in consumer subscriptions. The data terms also differ: consumer data trains by default, while paid API and Vertex AI data does not.
Alternative Tools
-
ChatGPT ChatGPT is a general-purpose AI assistant for writing, file and data analysis, web research, image generation, coding, and voice conversations. -
Qwen Chat Qwen Chat is Alibaba's free consumer AI assistant, offering 1M-token context, multimodal understanding, image and video generation, Deep Research, and web development tools, with optional paid consumer tiers for higher quotas. -
Claude Claude is Anthropic's AI assistant for writing, analysis, coding, web search, and delegating tasks that return finished documents, slides, spreadsheets, and designs. -
DeepSeek DeepSeek is an open-weight AI assistant with a free consumer chat app, a 1M-token context window, and a low-cost API supporting thinking mode and agent workflows. -
Grok Grok is xAI's AI assistant for chat, real-time search, image and video generation with Grok Imagine, voice conversations, file analysis, and coding and agentic tasks through Grok 4.5. -
Vibe (formerly Le Chat) Vibe (formerly Le Chat) is Mistral AI's unified assistant for chat, multi-step work tasks with connected apps, and agentic coding across web, IDE, and terminal, with a Free tier and paid Pro and Team plans.