White-Label GPU Cloud

Launch your own branded GPU cloud platform. Your brand, your pricing, your domain — our infrastructure.

What is White-Label?

Packet.ai White-Label lets you offer GPU cloud services under your own brand. Your customers interact with a complete, fully branded platform — your logo, your domain, your colors, your pricing — while Packet.ai handles all GPU infrastructure, provisioning, and management behind the scenes.

You purchase GPU hours at wholesale rates and set your own selling price. Customer payments go directly to your Stripe account. Packet.ai invoices you monthly for wholesale usage. No platform fees, no revenue shares.

What's Included

Complete Branded Website

Full Next.js app under your domain: dashboard, docs, status page, API docs

Embeddable Widgets

Pricing, checkout, auth, and GPU status widgets for your existing website

Tenant REST API

Customer management, pod provisioning, and widget endpoints

WHMCS Module

Full provisioning module with client area controls and admin panel

Stripe Integration

Direct payments to your Stripe account with AES-256-GCM encrypted keys

Custom Pricing

Set per-GPU hourly rates with any markup over wholesale

Preview URLs

Test your branded site instantly at {slug}.tenants.packet.ai

Sandbox Mode

Launch without Stripe keys to test branding and integration first

Complete Branded Website

Every tenant gets the full Packet.ai Next.js application served under their own domain with their own branding. This is not a widget or iframe — it is the complete platform, white-labeled.

Your customers see your brand name, logo, colors, and favicon on every page:

Customer dashboard — Deploy GPUs, manage running instances, view billing
Documentation pages (/docs/*) — Guides and reference material
Status page (/status) — Real-time GPU availability
API documentation (/api-docs) — Interactive API reference

Embeddable Widget System

In addition to the full branded website, you get embeddable widgets you can place on your existing marketing site. Widgets render inside Shadow DOM for complete style isolation and are CORS-enabled for any origin.

Install

Add this script to your HTML, replacing the domain with your tenant domain:

<script src="https://gpu.yourcompany.com/widget.js"></script>

Available Widgets

Use the data-packet-widget attribute to render different widget types:

<!-- GPU pricing table -->
<div data-packet-widget="pricing"></div>

<!-- Checkout / deploy flow -->
<div data-packet-widget="checkout"></div>

<!-- Signup / login form (email magic link) -->
<div data-packet-widget="auth"></div>

<!-- Real-time GPU availability -->
<div data-packet-widget="gpu-status"></div>

JavaScript API

Control widgets programmatically after the script loads:

// Initialize with options
PacketWidget.init({
  theme: "dark",        // "light" | "dark" | "auto"
  container: "#gpu-pricing"
});

// Listen for events
PacketWidget.on("deploy", (event) => {
  console.log("User deploying:", event.gpu);
});

PacketWidget.on("signup", (event) => {
  console.log("New signup:", event.email);
});

Tenant API

The Tenant API lets you manage customers and GPU pods programmatically. All requests authenticate with your tenant API key in the X-API-Key header.

Authentication

curl https://gpu.yourcompany.com/api/v1/tenant/customers \
  -H "X-API-Key: pak_your_api_key_here"

Tenant Management Endpoints

Method	Endpoint	Description
GET	/api/v1/tenant/customers	List all tenant customers
POST	/api/v1/tenant/customers	Create a new customer
GET	/api/v1/tenant/pods	List all customer pods
POST	/api/v1/tenant/pods	Provision a new GPU pod

Widget Endpoints

These endpoints power the embeddable widgets and are also available for direct use:

Method	Endpoint	Description
GET	/api/widget/config	Tenant branding configuration
GET	/api/widget/pricing	GPU pricing for the tenant
GET	/api/widget/gpu-status	Real-time GPU availability
POST	/api/widget/auth	Email magic-link signup/login

Example: Create a Customer

curl -X POST https://gpu.yourcompany.com/api/v1/tenant/customers \
  -H "X-API-Key: pak_your_api_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "email": "customer@example.com",
    "name": "Jane Smith"
  }'

Example: Provision a GPU Pod

curl -X POST https://gpu.yourcompany.com/api/v1/tenant/pods \
  -H "X-API-Key: pak_your_api_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "customer_id": "cust_abc123",
    "gpu_type": "rtx-pro-6000",
    "quantity": 1
  }'

WHMCS Module

A complete WHMCS provisioning module is available for hosting providers who manage their business through WHMCS. The module lives at whmcs-module/modules/servers/packetai/.

Features

Server configuration — API key and hostname setup in WHMCS
Per-product GPU selection — Map WHMCS products to GPU types (rtx-pro-6000, b200, h200, h100)
Client area — Pod status, controls (start/stop/restart), SSH terminal link, usage stats
Admin area — Pod details and dashboard links
Full lifecycle management — Create, suspend, unsuspend, and terminate pods automatically
Usage tracking — Daily usage collection cron for accurate billing

Installation

Download WHMCS Module (.zip)

Download the module using the button above
Extract and upload modules/servers/packetai/ to your WHMCS installation
Add a new server in WHMCS with your tenant API key and hostname
Create products and map each to a GPU type

Stripe Integration

Each tenant connects their own Stripe account. Customer payments go directly to you — Packet.ai never touches your customer revenue.

Provide your Stripe publishable key, secret key, and webhook secret
All keys are encrypted at rest with AES-256-GCM
Widget auth automatically creates Stripe customers in your account
Payments, invoices, and subscriptions are managed in your Stripe dashboard

Sandbox Mode: You can skip Stripe configuration entirely during testing. Enable Sandbox Mode in the onboarding wizard to launch your branded site without Stripe keys. Branding, pricing display, status page, and API docs all work normally — only checkout and deployment are disabled until you add Stripe keys.

Available GPUs

All GPUs are hosted in our US data centers with enterprise-grade infrastructure and 99.9% SLA.

GPU	VRAM	Wholesale
RTX PRO 6000 Blackwell	96 GB GDDR7	$0.66/hr
H200	141 GB HBM3e	$1.50/hr
B200	192 GB HBM3e	$2.25/hr
H100	80 GB HBM3	$2.49/hr

Requirements

Before you begin, make sure you have:

A domain or subdomain for your GPU cloud (e.g. gpu.yourcompany.com)

Brand assets: logo (SVG or PNG), favicon, and brand colors

A Stripe account (or use Sandbox Mode to start without one)

Setup Flow

Contact Packet.ai

Reach out to partners@packet.ai or use the contact form to request white-label access. Our team will set up your admin account within 24 hours.

Complete the Onboarding Wizard

Brand — Name, slug, logo, colors, favicon
Domains — Primary and additional domains
Stripe — API keys, or toggle Sandbox Mode to skip
GPUs — Select GPU types and set your selling prices
Support — Email, status page toggle, webhooks
Review — Confirm everything and create

Preview Your Site

As soon as the tenant is created, you can preview your fully branded site at the preview URL — no DNS changes needed:

https://{your-slug}.tenants.packet.ai

For example, if your slug is acme-gpu, your preview is at https://acme-gpu.tenants.packet.ai. The full site works at this URL: dashboard, docs, status page, and API docs.

Configure DNS

When you are ready to use your own domain, create a CNAME record:

gpu.yourcompany.com  CNAME  tenants.packet.ai

After DNS verification in the admin panel, an SSL certificate is automatically provisioned via Let's Encrypt and the Nginx configuration is generated and reloaded. Propagation typically takes 5-30 minutes.

Connect Stripe & Go Live

If you started in Sandbox Mode, add your Stripe keys in the admin panel to enable checkout and deployments. You will need:

Stripe publishable key (pk_live_)
Stripe secret key (sk_live_)
Stripe webhook secret (whsec_)

Tip: You can start with test keys (pk_test_ / sk_test_) to verify the Stripe integration before switching to live keys.

Pricing & Billing

White-Label uses a straightforward wholesale model:

Wholesale cost — What Packet.ai charges you per GPU per hour
Selling price — What you charge your customers (you decide)
Your margin — The difference is yours to keep

Payments from your customers go directly to your Stripe account. Packet.ai invoices you monthly for wholesale GPU usage. There are no platform fees or revenue shares beyond the wholesale cost.

Example: RTX PRO 6000 wholesale is $0.66/hr. You sell at $0.99/hr. Your margin is $0.33/hr per GPU (50% markup). With 10 GPUs running 24/7, that's approximately $2,376/mo in margin.

FAQ

What GPU types are available?

RTX PRO 6000 Blackwell (96 GB GDDR7), B200 (192 GB HBM3e), H200 (141 GB HBM3e), and H100 (80 GB HBM3).

Where are the GPUs located?

Our GPUs are hosted in US data centers with enterprise-grade infrastructure and 99.9% SLA.

Can I use my own domain?

Yes. Point your domain to tenants.packet.ai via CNAME. SSL certificates are provisioned automatically via Let's Encrypt.

Can I test before configuring DNS?

Yes. Every tenant gets a preview URL at {slug}.tenants.packet.ai that works immediately with the full branded site.

What is Sandbox Mode?

Sandbox Mode lets you create a tenant without Stripe keys. Your full branded site works — branding, pricing display, status page, API docs — but checkout and deployment are disabled until you add Stripe keys.

How does billing work?

Your customers pay you via Stripe. Packet.ai invoices you monthly for wholesale GPU usage. No upfront costs.

Can I customize the look and feel?

Yes. You control the brand name, logo, favicon, and brand colors. Every page of the platform renders with your branding.

What support do I get?

24/7 technical support from our engineering team. We handle all infrastructure issues. You handle your customer relationships.

Is there a minimum commitment?

No. Pay only for what your customers use. No minimum spend, no long-term contracts.

Do you support WHMCS?

Yes. A complete WHMCS provisioning module with client area controls, admin panel integration, and automated lifecycle management is included.

Ready to get started?

Contact our partnerships team and we'll have you set up within 24 hours.