> ## Documentation Index
> Fetch the complete documentation index at: https://docs.orq.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Deployment Options

> Compare Orq.ai hosting options from managed cloud to self-hosted on-premise, with VPC and sovereign deployment for enterprise compliance.

**Orq.ai** offers a range of deployment options to meet security, compliance, and data residency requirements.

## Overview

|                    | Cloud          | VPC                       | On-Premise            |
| ------------------ | -------------- | ------------------------- | --------------------- |
| **Setup**          | None           | Marketplace               | Helm chart            |
| **Infrastructure** | Orq.ai-managed | Customer AWS or Azure     | Customer-managed      |
| **Data residency** | EU default     | Customer VPC              | Customer environment  |
| **Air-gapped**     | No             | No                        | Yes                   |
| **Compliance**     | SOC 2, GDPR    | SOC 2 + customer controls | Full customer control |

## Cloud

The default deployment. Multi-tenanted, hosted in EU data centers on Google Cloud Platform. No setup required.

All customer data stays in the EU by default. SOC 2 Type 2 certified and GDPR compliant, with compliance continuously monitored through Vanta.

<Card title="EU Routing and Data Residency" icon="map-pin" href="/docs/enterprise/eu-regions-faq">
  How data residency works and which models offer EU inference.
</Card>

## VPC

**Orq.ai** runs inside the customer's own AWS or Azure account, fully isolated within the customer's network. Available through the AWS and Azure marketplaces.

The customer owns and controls the compute and storage. Platform data never leaves the customer's VPC.

<Card title="VPC Deployment" icon="cloud" href="/docs/enterprise/vpc-deployment">
  Subscribe and deploy through AWS or Azure marketplace.
</Card>

## On-Premise

**Orq.ai** deploys as a single Helm chart into a Kubernetes cluster, on any cloud provider or self-managed infrastructure. All components run inside the customer environment.

The platform operates fully air-gapped once container images are downloaded or cached in a local registry, as long as configured model endpoints are reachable from within the environment.

| Requirement | Details                                                          |
| ----------- | ---------------------------------------------------------------- |
| Kubernetes  | Version 1.28 or later                                            |
| Nodes       | At least 3 worker nodes, 4 vCPU and 16 GB RAM each               |
| Data stores | PostgreSQL 14-18, MongoDB-compatible, S3-protocol object storage |

<Card title="On-Prem Requirements" icon="server" href="/docs/enterprise/on-prem-requirements">
  Full infrastructure requirements, Helm chart configuration, and network access details.
</Card>

## Sovereign AI and Zero Data Retention

All deployment options support sovereign AI controls. The AI Gateway includes built-in filters for EU-only model routing and zero data retention (ZDR). PII masking is available at the platform level across all tiers.

<Card title="Sovereign AI and ZDR" icon="shield-check" href="/docs/enterprise/sovereign-ai">
  EU sovereignty, ZDR filter, PII anonymization, and no training on customer data.
</Card>

## Next Steps

For deployment assistance or enterprise pricing, contact [support@orq.ai](mailto:support@orq.ai) or reach out to an **Orq.ai** account manager.