BYOK: Using Your Own Provider Keys Behind One LLM API
Teams adopting a multi-model gateway usually hit the same procurement question: "We already have contracts and negotiated prices with our model providers — why would we re-buy inference through a middleman?"
Bring Your Own Key (BYOK) is the answer to exactly that. You keep your provider accounts; the gateway keeps the integration surface.
How it works
With BYOK on BoostRail, an organization stores its own provider API keys with the platform. When a request targets a model covered by one of those keys, the gateway serves it with your key against the provider's official endpoint. That means:
- Inference charges land on your own provider account, at your contract prices — not on platform credit.
- Request data stays within your existing provider agreements, which matters where procurement or compliance requires a direct provider relationship.
- Your application never touches the stored keys — they are backend inventory used by the routing layer.
What you keep from the gateway side: the single OpenAI-compatible endpoint, one key for your application, the unified catalog, health-scored routing, and automatic platform fallback — if your upstream key hits quota or errors, requests can fall back to platform lines instead of failing.
What it costs
The free plan includes 1,000,000 BYOK requests per month with no top-up required. Paid plans raise the monthly quota and widen the provider allowlist. Since inference itself is billed by your provider, the platform side is a software fee on requests — not a markup on tokens.
Zero lock-in, by construction
Lock-in is the standard objection to gateways, and BYOK inverts it: the provider keys are yours, and your application talks to an OpenAI-compatible endpoint. Migration off the gateway is pointing base_url back at a provider's official endpoint. There is no proprietary SDK, no custom protocol, and no data you cannot walk away with.
When BYOK covers a model — and when it doesn't
BYOK routing requires two things: the provider is on the supported list, and the model has an official-endpoint mapping. Models sourced through aggregated channels, or from providers without a public official API, run on platform lines only. Coverage grows with the catalog; the current provider allowlist and per-plan quotas are on the BYOK page.
Typical setups
- Contract-price teams: inference on negotiated provider prices, gateway for routing/observability, platform fallback for spillover.
- Compliance-bound teams: requests to sensitive models pinned to their own provider agreement, everything else on platform lines.
- Cost-split teams: frontier models via their own keys (where contracts are), long-tail models on platform credit — one bill for the software fee, provider bills unchanged.
An API key and the BYOK setup flow are at app.boostrail.com; a step-by-step guide lives in the console.