Short answer. Gemini Enterprise is Google's employee AI app, and the Gemini Enterprise Agent Platform is the Google Cloud layer for building and governing agents. xpander is a vendor-neutral AI agent control plane. It runs Claude Code, Codex, OpenCode and agents you describe in plain language, on the models your admins allow. One set of access, approval and budget rules covers all of them. It runs in xpander's cloud, in your own Kubernetes cluster, or air-gapped. Choose Gemini Enterprise for a Google-first company. Choose xpander when agents and models come from many vendors, or when agents must run in your own environment.
What each product is
Google describes Gemini Enterprise as "an intranet search, AI assistant, and agentic platform", and the Agent Platform as a platform to "build, scale, govern, and optimize enterprise-grade agents grounded in your enterprise data." Agent Platform is the evolution of Vertex AI: "all Vertex AI services and roadmap evolutions will be delivered exclusively through the Agent Platform."
xpander is one platform. Its parts share the same sign-in, rules and records. The agent runtime runs the agents. The AI Gateway carries every model call. The MCP Gateway gives each person's AI tools their approved tools. Employees use agents on the web, in the desktop app, in Slack and in Teams.
Where Google is strong
Model and infrastructure depth. Agent Platform offers "more than 200" models through Model Garden, including Anthropic's Claude.
Network-level agent governance. Agent Gateway blocks every connection by default "unless an explicit IAM policy grants access", inspects MCP calls and governs A2A calls, and gives each agent its own identity.
Data controls. Gemini Enterprise supports VPC Service Controls, customer-managed encryption keys and data residency, and Agent Platform applies Model Armor policies.
Search across company data. Gemini Enterprise answers from more than 100 connected sources, following each person's existing permissions.
Side by side
xpander | Gemini Enterprise + Agent Platform | |
|---|---|---|
What it is | A vendor-neutral control plane that runs and governs agents | Google's employee AI app plus its Google Cloud agent platform |
Where it runs | xpander cloud, your own Kubernetes cluster in any cloud or on-prem, or air-gapped | Google Cloud. Agent Gateway can govern agents on Kubernetes clusters outside Google Cloud, including your own data centers (limited preview) |
Self-hosted or air-gapped | Yes. Air-gapped installs run every component in your network, with no connection to xpander cloud | Not documented for the Gemini Enterprise app or Agent Platform. Gemini models run on Google Distributed Cloud air-gapped (generally available since August 2025). Enterprise search with Gemini on GDC air-gapped is in preview (October 20, 2026) |
Models | The models your admins allow, including open-weight models on your own GPUs | Gemini models. An admin can turn on Claude Opus 5.5 and Claude Sonnet 5.5 for the coding tools. Wider third-party model choice is in early access |
Agents it runs | Claude Code, Codex, OpenCode and custom agents | Agents on Agent Runtime, A2A agents, Dialogflow agents and no-code agents from Workflow Builder |
Coding agents | Claude Code, Codex and OpenCode as governed agents | Google Antigravity, Antigravity for IDEs and Android Studio are the bundled coding tools. Google documents routing Codex through Agent Platform (private preview) |
Spend control | Budgets in credits that warn or block per organization, group, member, API key, agent, turn or end user | A spend cap per project that pauses paid usage, and a per-user quota limit. Seat credits apply only to Gemini models |
Record | An audit log of sign-ins, setting changes and every tool call, plus a log of every model call. The audit log exports to your OpenTelemetry collector or storage bucket | Cloud Logging and Cloud Trace |
Pricing | xpander cloud: credits for the work agents do, with unlimited seats. Self-hosted: an annual license, from 50 agents, with your own model keys | Per-seat editions plus pay-as-you-go usage. Agent Platform bills tokens, compute, memory and storage |
The key differences
One cloud or any. Gemini Enterprise and Agent Platform run on Google Cloud, and identity, policy and logs live in Google Cloud IAM, Cloud Logging and Cloud Trace. Agent Gateway can govern agents on Kubernetes clusters outside Google Cloud, including your own data centers (limited preview). xpander runs in its own cloud, in your cluster on any cloud, or air-gapped.
Whose agents. Gemini Enterprise registers agents hosted on Agent Runtime, A2A agents, Dialogflow agents and no-code agents, and bundles Google Antigravity for coding. xpander runs the coding agents teams already use, Claude Code, Codex and OpenCode, next to custom agents, under one policy.
Whose models. Seat credits in Gemini Enterprise "apply strictly to first-party Gemini models", and partner models are billed as extra usage. xpander puts the models your admins allow, commercial or self-hosted, behind one gateway with the same budgets.
How a coding agent runs under xpander
An admin adds providers and models in Platform > AI governance > Models.
A person picks Claude Code, Codex or OpenCode and a model in the Harness and model picker.
The agent calls the AI Gateway with an xpander key or session token, never a provider key.
A call over a blocking budget is refused and says so. The call is in Logs, and the action is in the audit log.
How to choose
If your situation is... | Choose |
|---|---|
The company runs on Google Workspace and Google Cloud | Gemini Enterprise |
You want company search with Gemini for every employee | Gemini Enterprise |
Teams use Claude Code, Codex or several model vendors | xpander |
Agents, not only models, must run on-prem, on another cloud or air-gapped | xpander |
You need budgets per person, group and agent across every model you allow | xpander |
Using both
Gemini models can run through the xpander AI Gateway like any other model, so a company can keep Gemini Enterprise for search and use xpander to run and govern agents across vendors.
Frequently asked questions
Is xpander an alternative to Gemini Enterprise?
For companies that need a vendor-neutral control plane, yes. Gemini Enterprise is Google's employee AI app and agent platform on Google Cloud. xpander runs and governs Claude Code, Codex, OpenCode and custom agents, on the models your admins allow, in xpander's cloud, your Kubernetes or air-gapped.
Can Gemini Enterprise run air-gapped?
Google does not document an air-gapped version of the Gemini Enterprise app or Agent Platform. Gemini models run on Google Distributed Cloud air-gapped, generally available since August 2025, and enterprise search with Gemini on GDC air-gapped is in preview since October 20, 2026. On xpander, air-gapped installs run every component in your network, with no connection to xpander cloud.
Can I use Gemini models in xpander?
Yes. xpander's AI Gateway works with commercial models such as Gemini, Claude and GPT, and with open-weight models you host yourself.
Does Gemini Enterprise govern Claude Code or Codex?
Google documents routing Codex CLI and the Codex desktop app through Agent Platform, with sign-in through Google Cloud IAM (private preview). Google's documentation does not describe Claude Code. In xpander, Claude Code, Codex and OpenCode run as governed agents.
Facts checked 18 Nov 2026.
Sources
xpander: Deployment options, AI Gateway, Budgets, Approvals, Audit log, Pricing
