Resource guide

LLM Usage Metering for GPT-compatible API operations

Understand how LLM usage metering works when API usage, credits, routing, and customer workspaces live in one operating layer.

Start tracking usage View setup guide

OpenAI-compatible setup with transparent usage logs.

OpenAI-compatiblePrepaid creditsUsage logsReseller-ready

Lower-cost GPT-compatible routesPrepaid spend controlVisible per-request deductions

Cost visibleCredits, deductions, and runway stay readable.

Routes observableHealth, fallback, and request logs stay close to the API.

Keys controlledCreate, limit, and revoke keys without exposing provider credentials.

How it works

Change the endpoint, then keep operations visible

Create a Rlab key, set the OpenAI-compatible base URL, send one request, then inspect credits, model usage, and logs from the console. For reseller flows, customer workspaces and attribution stay attached to signup and recharge paths.

LLM Usage Metering for GPT-compatible API operations diagram showing Rlab API credits, routing, and usage logs. — A product diagram explaining how Rlab connects GPT-compatible API access with prepaid credits and operational logs.

Built for practical teams

Developers

Ship with an OpenAI-compatible endpoint and avoid rebuilding billing later.

Teams

Keep API keys, credits, and usage logs in one operating layer.

AI agencies

Give customers workspaces and usage visibility without spreadsheets.

Resellers

Connect referral attribution, prepaid recharge, and customer usage.

Operating playbook

From first request to a cleaner API business loop

LLM Usage Metering for GPT-compatible API operations should not stop at a cheaper endpoint. The useful version is an operating loop: create a key, send requests through a GPT-compatible route, meter usage, deduct prepaid credits, and keep a ledger that the developer, operator, and customer can all read without asking for raw provider credentials.

For teams, every integration can have its own key and every key can carry a business meaning: production app, internal agent, customer workspace, reseller account, or test environment. When usage grows, the logs are already shaped for review instead of becoming a surprise invoice at the end of the month.

For resellers and agencies, a signup link or customer invite should keep attribution attached, the customer workspace should keep its balance, and the recharge record should connect back to the same operating account. Rlab SEO pages link into the existing signup path with additive intent and UTM parameters.

The result is a product surface that developers can adopt quickly while business teams keep cost, route quality, and customer billing visible.

Attribution stays intactSEO intent and UTM parameters are additive, while invite, channel_user_invite, referrer, customer_type, and type remain protected.

Billing stays readablePrepaid credit deductions, usage logs, and ledger records keep the money trail close to the API call.

Routes stay observablePriority, fallback, health, and request logs help operators explain quality instead of guessing.

Where Rlab fits

Generic gatewayGood for routing, often thin on customer billing.

Model marketplaceUseful for discovery, less focused on reseller ledgers.

RlabOpenAI-compatible access plus prepaid credits, logs, workspaces, and attribution.

FAQ

Is Rlab OpenAI-compatible?

Yes. Rlab is designed around an OpenAI-compatible API surface so many apps can start by changing the base URL and API key.

How do prepaid credits work?

Teams recharge credits first, API calls are metered, and deductions are written to usage and ledger records.

Can I track usage by key?

Yes. The product model includes API keys, request logs, model usage, and billing history for operational review.

Does this replace my app code?

No. For compatible clients, the goal is to keep integration small and add billing and routing controls around it.

Can resellers use it?

Yes. Customer workspaces, prepaid balances, attribution, and recharge flows are part of the reseller operating model.