# White-label LLM API: your brand on top, a managed upstream underneath

> Markdown mirror of https://router.one/white-label-llm-api for AI assistants and crawlers. Router One is an OpenAI-compatible LLM API gateway.

If you are searching for a white-label LLM API, here is the honest boundary first: Router One does not sell a turnkey white-label storefront, sub-accounts, or rebrandable dashboards. What it provides is the layer under your brand — one OpenAI-compatible endpoint for 40+ models, multiple API keys with per-key budgets (maxSpend) and rate limits, and per-request usage data so your own portal can bill your own customers. The pay-as-you-go token line uses posted model rates, with checkout-visible FX/channel fees kept separate; the brand, the pricing, and the customer relationship stay entirely yours.

## What “white-label” means here

Router One has no relationship with your end customers — no signup they see, no branding in your product, no emails to your users. Your customers interact with the portal, app, or relay you run; calls to Router One happen from your backend with your API key. That is the whole point of the split: a turnkey storefront rents you a brand surface on someone else's platform, while building on a managed upstream keeps the customer, the data, and the margin in your hands.

## You bring the brand layer

- Your portal, app, or relay software (one-api / new-api) under your own name
- Your customer pricing, and the margin you charge over upstream cost
- Your billing, invoicing, and customer accounts
- Your signup, support, and terms with your customers

## Router One provides the upstream

- One OpenAI-compatible endpoint for 40+ models (GPT, Claude, Gemini, Grok, DeepSeek, GLM, Kimi)
- Multiple API keys per account, each with per-key budgets (maxSpend) and rate limits
- Per-request usage data — model, tokens, cost, latency, status — to drive your billing
- A pay-as-you-go token line at posted model rates, with checkout-visible FX/channel fees kept separate

## Branded service on Router One vs a turnkey white-label storefront

| What you control | Built on Router One | Turnkey white-label storefront |
| --- | --- | --- |
| Your brand and portal | Yours entirely — Router One never appears in front of your customers | Rented surface on the platform's product, on its terms |
| Customer data and relationship | Live in your systems; Router One only sees your backend's API calls | Often live inside the platform |
| Upstream transparency | Per-request trace: model, tokens, cost, latency, status | Varies by platform |
| Per-customer spend control | One key per customer or tier, each with maxSpend, rate limit, and TPM cap | Coarse or plan-level only |
| Time to a hosted branded UI | You build or bring one — one-api / new-api work out of the box | Faster if a hosted storefront is all you need |
| Pricing | Posted per-model rates; your customer margin is set in your portal | Bundled pricing, often with the platform's cut inside |

## Per-key budgets do the multi-customer metering

Issue one API key per customer, tier, or environment. Each key carries its own maxSpend, request rate limit, and tokens-per-minute cap, so a single customer can never drain your whole balance — and the per-request usage data tells your billing system exactly what that key consumed. Your customers never hold your master credentials; keys are the isolation unit.

## What Router One does not provide (so you can plan for it)

- White-label branding or a rebrandable dashboard
- Sub-accounts / child accounts for your customers
- Configurable markup multipliers — your margin lives in your own portal
- Wholesale or volume discount tiers

## FAQ

### Does Router One offer a white-label LLM API?

Not as a turnkey product. Router One is the managed upstream you build a branded service on: one OpenAI-compatible endpoint for 40+ models, per-key budgets and rate limits, and per-request usage data. The white-label part — portal, brand, customer billing — is the part you own.

### Will my customers ever see Router One?

Router One has no relationship with your end customers: no signup they touch, no branding in your product, no emails to your users. What your customers see is decided entirely by the product you build on top. Requests to Router One are backend calls made with your API key.

### How do I meter and cap each customer?

One API key per customer or tier. Each key has its own maxSpend budget, request rate limit, and tokens-per-minute cap, and every request is traced with model, tokens, cost, latency, and status — so your portal can bill from real usage data and no single customer can run away with spend.

### Does Router One take a cut of my customer revenue?

No revenue share. You pay the pay-as-you-go token line at posted model rates, with checkout-visible FX/channel fees kept separate. Whatever you charge your customers on top is yours.

### Can I get sub-accounts for my customers?

Not today. Keys are the isolation unit: issue one per customer or tier, each with its own budget and limits. Sub-accounts, organization roles, and rebrandable dashboards are not provided — that layer belongs to your product.

## See also

- LLM API reseller: https://router.one/llm-api-reseller
- API relay station: https://router.one/llm-api-relay-station
- LLM API gateway: https://router.one/llm-api-gateway
- LLM observability: https://router.one/llm-observability
- Canonical page: https://router.one/white-label-llm-api
- Models and per-model token rates: https://router.one/models (markdown: https://router.one/models.md)
- Pricing: https://router.one/pricing
- API docs (markdown): https://router.one/docs.md
- Company facts: https://router.one/facts/company.md