Gemini Enterprise Pricing in 2026: Business, Standard and Plus Editions vs Private AI

Gemini Enterprise pricing is $21 per user per month for the Business edition on an annual prepaid term, $30 for Standard and $50 for Plus on one-year terms, or $25, $35 and $60 month to month, according to Google Cloud's published SKU price list (verified 14 October 2026). Use beyond each edition's pooled quotas is billed at Agent Platform rates, and indexed data above the pool costs $5 per GiB per month.

Google’s product pages say little about Gemini Enterprise prices. The main page, now marketed as Gemini for business, points buyers to sales and a 30-day Plus trial, and the Business edition page shows a starting price of $21 per seat. The Standard and Plus figures sit in Google Cloud’s SKU catalogue, the published list of Google Cloud SKUs and prices. Every figure below comes from that catalogue, Google’s Gemini Enterprise documentation, the Workspace pricing page and the Agent Platform price list, checked on 14 October 2026.

Three products share the Gemini name at work. Gemini Enterprise is a standalone agent and search app with per-user editions. Gemini in Google Workspace is built into Gmail, Docs and the other apps. Developers call Gemini models through Gemini Enterprise Agent Platform, formerly Vertex AI, which bills per token and is bought separately; our comparison with Google’s agent platform covers that side.

Gemini Enterprise pricing plans by edition

Subscriptions run monthly or for a year and can renew automatically; cancelling means switching off auto-renewal, and charges continue to the end of the term. The Cloud console caps seat purchases at 25 per billing account for self-serve and resold accounts and at 1,000 for invoiced accounts, so larger orders go through Google’s sales team or a partner. Google’s edition comparison allows Business for 1 to 500 users, although its FAQ and marketing page pitch the edition at teams of up to 300.

Gemini Enterprise pricing per user per month in euros and rupees

The euro and rupee figures are Google’s own list prices in those currencies, and its SKU page notes that it will not give notice when non-dollar prices rise because of currency movements. The rupee list also carries a Standard Emerging subscription at ₹760 per user per month on a one-year term ($8.50 in the dollar list). Google’s quota table describes it as an emerging-market edition with 40 assistant queries a day, and its public pages do not say who qualifies.

Gemini Business pricing compared with Standard and Plus

Google’s FAQ spells out what the compliance row means for Standard and Plus: data residency, Google Cloud audit logging, customer-managed encryption keys, VPC Service Controls, and FedRAMP and HIPAA coverage, with sign-in through Google or third-party identity providers. Business is aimed at small teams that want no IT setup, and Google also sells it through Google Workspace subscriptions via its sales team and partners. If enterprise search across many systems is the goal, our enterprise search comparison sets Gemini Enterprise beside eight other tools.

Quotas, overages and pay-as-you-go charges

Every Gemini Enterprise seat brings daily quotas, and the quotas are pooled across all seats of the same edition in a project and location. One hundred Standard seats therefore share 16,000 assistant queries a day, and a heavy user can draw beyond 160 while the pool lasts. Pools reset at midnight Pacific Time, except the AI developer tools credit, which runs on a seven-day cycle. Storage and indexing is pooled more widely still, across every edition in the project and location.

When a pool is empty, Google restricts that feature unless an administrator has switched on overages. Overages need an invoiced Cloud Billing account and a paid, non-trial subscription. Storage beyond the pool costs $5 per GiB per month, prorated for the time the data is held. Other features, such as assistant queries, Deep Research and media generation, are charged at Agent Platform prices, and a variable support fee may apply. The Pay-as-you-go edition charges everything this way.

The Business edition has no overage option at all. When its pooled quota is gone for the day, Google’s help centre says the choices are more seats or an upgrade to Standard or Plus.

Gemini in Google Workspace plans

Workspace business plans include Gemini without an add-on, at different depths. Google shows Workspace prices in the visitor’s local currency, and from an EU address on 14 October 2026 its pricing page listed:

These prices exclude Google’s new-customer offers, which on the day reached up to 50% off for the first three months. Google’s usage-limits page sets monthly AI limits per user that rise from Business Starter to the Enterprise plans, and it describes an AI Expanded Access add-on to raise them for chosen users without listing a price.

The two products overlap without replacing each other. Workspace’s Gemini works inside Google’s own apps; Gemini Enterprise searches and runs agents across connected systems such as Microsoft 365, Confluence, Jira and ServiceNow. A Workspace customer who wants both pays for both.

Gemini Enterprise API pricing on Agent Platform

A Gemini Enterprise seat does not include developer access to Gemini models. Google’s FAQ is explicit that Gemini Enterprise Agent Platform, the evolution of Vertex AI, is bought separately on a Google Cloud account with pay-as-you-go billing. Its list prices on 14 October 2026 included:

Google delivers the promotional rates as 50% credits back on net spend. Non-global endpoints cost 10% more, and grounding with Google Search is free for the first 5,000 queries a month and then $14 per 1,000. Overages in the Gemini Enterprise app are charged from this same price list.

Gemini Enterprise cost for 250 and 1,000 users

How to read it:

• Pools grow with seats. 250 Standard seats bring 7,500 GiB of pooled storage and 40,000 assistant queries a day; 1,000 bring 30,000 GiB and 160,000 queries. Every 100 GiB indexed beyond the pool adds $500 a month.
• A 1,000-seat order needs sales or an invoiced account, because the console stops self-serve accounts at 25 seats.
• Monthly terms cost 17% to 20% more than one-year terms across the three editions.

What the Gemini seat price leaves out

• Overages, once enabled, for queries, Deep Research and media generation at Agent Platform prices.
• Index storage above the pooled allowance, at $5 per GiB per month.
• Developer work. Agents built in code, model API calls and grounding fees on Agent Platform.
• Workspace licences, if staff also need Gemini inside Gmail and Docs.
• Rollout. Connector set-up, permission reviews and a pilot before the full order.

The same lines are worked through for Claude Enterprise and Microsoft 365 Copilot, and for OpenAI in our ChatGPT seat-cost breakdown.

What a private deployment costs instead

Gemini Enterprise prices capacity twice over: once in seats, and again in the quotas and storage pools those seats carry. A private deployment prices it in equipment and people. The lines to expect:

• GPU servers sized for the peak number of requests running at once; the user count matters only through that peak.
• Index storage on your own disks, so a large document corpus becomes a hardware purchase rather than a monthly fee per GiB.
• Electricity and cooling for the life of the hardware.
• An operator’s time for upgrades, monitoring and support.
• Licences and support for the platform software and the models you choose.
• Utilisation. The busier the hardware, the lower the cost of each answer.

A fair comparison sets three years of Gemini seats, overages and Agent Platform usage against three years of those lines, using query volumes from a pilot. It also leaves out data that may not be processed in Google Cloud, which no edition can serve. Our breakeven analysis for on-premises models works through hardware, energy and staffing, the pricing-model guide explains how committed and metered charges behave, the CFO’s budgeting guide frames the spend for finance, and the AI savings calculator lets you test seat counts.

How VDF AI fits

VDF AI overlaps with Gemini Enterprise on agents and on answers drawn from company documents, but it runs on infrastructure you control: governed agents and workflows, private RAG over company data, chat, audit logs and role-based access. In a self-hosted deployment, documents are indexed on your own storage, so the size of the index is a capacity decision rather than a monthly fee.

VDF AI Router lists Ollama and custom on-premises deployments as model targets beside the cloud models your policy approves, and its air-gap mode confines routing to local models. On-premises licensing is by Seat Bundle, with API Expansion Packs for programmatic use, and the licence does not meter model tokens. Teams that prefer a hosted service can use VDF AI Cloud, from a free single-user plan through Starter at $20 a month for up to three users and Professional at $50 per seat a month for up to 20 seats, with Enterprise Cloud through our sales team. Details are on the pricing page.