Two deployment paths

We can host RestrictAIChat—or help you run it in your environment.

Choose managed hosting for the shortest path to production, or deploy the Private Edition on infrastructure you control. Our team can support planning, installation, configuration, verification, and handover.

Choose your operating model

One governed workspace, two ways to operate.

Managed Hosting and Private Edition deliver the same governed workspace through different hosting and credential models.

Hosted service

Managed hosting

We operate the agreed application environment and help configure your governed workspace. Application data is stored in the agreed hosted environment. Data region, retention, support access, subprocessors, backup schedule, monitoring, upgrades, and support commitments are governed by the executed agreement.

  • Platform model access is available by default; optional BYOK can be reviewed
  • Your administrators control users, approved models, policies, retention, and budgets
  • Operational scope and service levels are agreed in the order form

Best for: Teams that want the shortest path to production without operating the application host.

Customer-hosted / on-premises

Private Edition with implementation support

Run the Private Edition in your VM, VPC, cloud account, or on-premises infrastructure. You control stored application data and model credentials; approved prompts are sent to the model providers you configure, and responses are returned by them, under their terms. We help your team move from architecture review to a verified handover.

  • Your infrastructure, storage, backups, keys, network, and provider accounts stay under your control
  • Guided installation plus workspace, identity, policy, access, and provider configuration
  • Health, readiness, login, audit-chain, backup, and recovery checks before handover

Best for: Organizations that require infrastructure, stored application data, and model credentials to remain under their control.

Implementation support

A defined path from requirements to production.

Managed Hosting and Private Edition use the same four-stage launch sequence.

  1. Discover

    Confirm users, regions, data boundaries, identity, network, providers, retention, and support requirements.

  2. Design

    Select the deployment model and document architecture, rollout plan, and acceptance criteria.

  3. Implement

    Prepare the environment, configure the workspace, connect approved services, and apply access and policy defaults.

  4. Verify and launch

    Test health, login, policy outcomes, audit evidence, backup and recovery, then hand over the operating runbook.

Technical baseline

Private Edition technical specifications.

These are the current supported starting conditions for customer-hosted deployment. Managed-hosting infrastructure is operated to an agreed service scope.

AreaPrivate Edition baseline
RuntimeDocker Engine with Compose and Node.js 24 inside the application image on a Linux amd64 or arm64 host with Landlock ABI 3
Application topologyOne long-running application container with a Caddy reverse proxy on a VM or equivalent host; edge, serverless and Kubernetes runtimes are excluded
Starting size2 vCPU, 4 GB RAM and 10 GB free disk, sized and load-tested for concurrency, retention and document volume
NetworkDomain, OpenSSL and HTTPS through inbound TCP 80/443 with Caddy or a dedicated outbound Cloudflare Tunnel bound to loopback; egress is limited to configured services
Data and backupA local persistent /data volume holds SQLite WAL, uploads and CLI credential homes; NFS is excluded, and backup verification includes deployment keys and restoration
File processingTesseract OCR with English, Traditional Chinese, Simplified Chinese and Japanese language packs, plus Poppler
IdentityLocal and directory passwords, TOTP, Google Workspace and Microsoft Entra ID sign-in, roles, groups, and SCIM Bearer create, read, deactivate and delete
Model access and billingCustomer-supplied API credentials or supported CLI subscriptions billed directly by the provider, with no RestrictAIChat usage markup
Implementation acceptanceInstaller preflight plus migration, health, readiness, login, audit-chain, backup and recovery checks

Regional service routes

Implementation partners by market

For managed hosting, Private Edition, or implementation support, contact the implementation partner published for your market. Each card names that partner, the market it covers, and its direct contact route.

Email reaches the published partner for that market directly. The inquiry form on this page is handled by RestrictAIChat.

Plan the operating model

Tell us where it needs to run.

Share your region, identity, network, model provider, retention, and support requirements. We will recommend a deployment path and implementation scope.

Plan your deployment