Introduction
Get started with nRouter: the enterprise LLM gateway delivering unified access, automated failover, strict budget controls, and cost tracking with one API key.
nRouter is an enterprise LLM gateway that unifies access to top AI models with real-time budget enforcement, in-path guardrails, and automated failover through a single OpenAI-compatible API.
- One key, one bill: Route across leading providers without managing individual vendor keys or accounts.
- In-path governance: Enforce credit budgets, RPM/TPM rate limits, and guardrails in real time before requests reach providers.
- 0% platform fee: Pay as you go pricing settles at the served model's exact list price with zero markup.
- OpenAI-compatible: Connect instantly by setting the base URL to
https://api.nrouter.ai/v1and supplying ansk-nrouter-*key. - Dynamic catalog: Query available models via the live Models API or explore the browsable Models directory.
See Why nRouter for architectural comparisons.
Request flow
What happens inside nRouter when your app calls a model.
Your application
OpenAI SDK · sk-nrouter-…
nRouter Rust Gateway
direct customer inference
- Virtual-key auth · tenancy
- Route · IP · lifecycle policy
- Budgets · rate limits
- Pre-call guardrails
- Cortex sidecar
- Reserve credits
- Provider routing
Providers
- OpenAI
- Anthropic
- Vertex
- Azure
- Alibaba
response ↩ read cost · settle credits · log · stream back
Onboarding
From sign-up to your first API call in six steps.
Sign up
Email & password
Verify email
magic-link or pwd
Activate
Verify phone or add $5+
Set up team
Default auto-created
Create key
sk-nrouter-… (once)
First call
x-nr-* headers
One org per user · Default team auto-created · key shown once
Architecture
The services in the request path and where data lives.
Your application
nRouter Rust Gateway
Policy · tenancy · money
- Virtual-key auth · Guardrails
- Credits · Budgets · Rate limits
- Tenant derived from key row
Provider routing
- Provider transforms · Fallbacks
- Cost settle · Spend logging
- Canonical Postgres store
Providers
OpenAI · Anthropic · Google Vertex AI · Azure Foundry · Alibaba
The services in the request path and where data lives.
Quick start
- Sign up at nrouter.ai/signup and generate a key at
/[org]/keys. - Send requests to
https://api.nrouter.ai/v1/chat/completions:
curl https://api.nrouter.ai/v1/chat/completions \
-H "Authorization: Bearer $NROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "gemini-2.5-flash", "messages": [{"role": "user", "content": "Hello!"}]}'Every request verifies keys, evaluates guardrails, routes traffic, and returns upstream output with x-nr-* telemetry headers at exact list price (0% fee). See Pricing.
Next Steps
- Quick Start — Make your first API call in under 2 minutes
- Authentication — Learn how API keys and auth work
- Chat Completions — Explore the Chat Completions API
- Python SDK — Get started with the Python SDK