Every model at its list price.

One balance and one API key, with no subscription required. Each request costs exactly what the model's provider charges, with a 0% platform fee.

  1. Add credits

    From $5. Card fees are on us.

  2. Call any model

    One API key for every provider.

  3. Pay per request

    At list price, with 0% markup.

Plans

Your keys, our gateway. Everything else on credits.

Pay as you go — $0/mo

No commitment. Buy credits and pay only for what you use.

$5to start
0%platform feeCard fees on us
Credit amount
Get started

Buy credits, use any model. 0% fee.

Secure checkoutPowered by Stripe

Card details go straight to Stripe. We never see or store them.

  • $5 minimum to start · pay as you go
  • 0% platform fee on credit purchases
  • 200 RPM · 5 concurrent requests
  • 7-day request log retention
  • Standard model access at flat list price

Bring your own keys

Use your own model keys on our gateway. Every other model runs on credits.

$20/mo
0%platform feeAlways 0% on subscriptions
Choose Bring your own keys

Use your OpenAI & Anthropic keys. Other models use credits. 50M tokens/mo included.

  • 50M tokens a month on your own keys
  • Every other model on credits
  • Higher rate limits than Pay as you go
  • Every feature on every plan

Enterprise

Custom volume & SLA

Custom credit pricing, dedicated capacity, custom SLA, and premium support.

Custom rate limitsDedicated throughput optionsVolume pricing and custom termsSecurity and compliance review
Custom
Tailored to your volume
Contact sales

Compare plans

Rate limits, data retention and features, side by side.

Scroll sideways to see every plan.

Plan comparison: pricing, throughput, features, and support across all tiers
Plan detailPay as you goBring your own keysEnterprise
Pricing
Platform fee0%0%Custom
Price$5+ credits$20/moCustom
nrouter/auto accessIncluded (from credits)Included (own keys + credits)Custom
Bring your own keys (OpenAI & Anthropic)—50M tokens/mo includedCustom
Billing cadencePay-as-you-goMonthlyCustom
SupportStandard / DocsStandard / DocsDedicated + custom SLA
Rate limits
Requests per minute (RPM)200500Custom
Tokens per minute (TPM)200,000500,000Custom
Requests per second (RPS)1025Custom
Requests per day (RPD)10,00050,000Custom
Concurrent requests510Custom
Max output tokens (ceiling)StandardStandardCustom
Data retention
Request & spend logs7 days7 daysContract (30+ days)
Platform audit logs18 months18 months18 months
Auth events12 months12 months12 months
Financial ledger7 years7 years7 years
Included on every plan
Guardrails — safety, PII, injection defense
Prompt templates, versioning & optimization
A/B testing with deterministic splitting
Spend analytics, observability & budgets
Custom branding & data-retention policies
Live models across every provider
Unlimited teams, members & keysUnlimitedUnlimitedUnlimited
SSO / SAML
Role-based access control (RBAC)
Audit logging
100% of every credit purchase becomes credits
Price-match guarantee
Enterprise infrastructure
Dedicated capacity & provisioned throughput——
Custom SLA——
Dedicated technical contact & named CSM——
BYO cloud / VPC deployment——
Regional data residency pinning——
Custom MSA & security review——
Estimate your spend

FAQs

What is the usage allowance?

The usage allowance is a monthly limit on requests routed to the nrouter/auto model for existing subscribers. It is not a credit balance and has no cash value. Unused allowance does not roll over and expires when the window resets. On Pay as you go, nrouter/auto is paid directly from credits at the served model's list price.

What is nrouter/auto?

nrouter/auto is automatic routing: send it as the model and nRouter chooses an eligible model from the published chat models your organization allows, without you needing to specify a provider. Best for the task (the default) uses smart model selection when it is available; Lowest cost orders the models fit for the task by their list rate. The response names the model that answered in x-nr-model. It is available to Pay as you go customers and paid from credits at the served model's list price.

Can I use my own OpenAI or Anthropic keys?

Yes. nRouter manages provider keys for you on credits, or you can bring your own OpenAI and Anthropic keys on the Bring your own keys plan ($20/mo, 50M tokens/mo included for text chat). Other models keep running on credits. Developers change nothing: use the same sk-nrouter key, endpoint, and SDK. Each response says who paid in the x-nr-funding-source header (byok or credits).

What happens when I hit a limit?

Pay as you go rate limits are 200 RPM, 200k TPM, and 5 concurrent requests. If you exhaust your credit balance, requests pause until you purchase more credits (credits never go negative). Existing subscribers with a usage allowance who reach an 8-hour window limit must wait for the window to reset or fund extra usage from credits.

Is capacity guaranteed?

Capacity is shared across all customers in your tier. During peak times, a request can wait or be asked to retry. Enterprise plans include dedicated capacity and custom SLAs.

How long is data retained?

Request logs are retained for up to 30 days, depending on your plan; prompt and response text are never stored. Platform audit logs are kept for 18 months, auth events for 12 months, and financial records for 7 years.

Do you offer refunds?

Credits are non-refundable once consumed by API usage. Unused credit balances may be refunded upon account closure, subject to our refund policy. Every refund request is reviewed by a person before it is issued.

How is the platform fee calculated on Pay as you go?

Pay as you go carries a 0% platform fee: $100 of credits costs $100.00, and the $5 minimum purchase costs $5.00. You receive 100% of the credits you buy, and usage settles at the served model's list price.

Who pays the card processing fee?

We do. nRouter covers the card processing fee, so the $5 purchase charges your card $5.00 and adds $5.00 of credits. Payments are processed by Stripe: your card details go straight to Stripe and never reach nRouter's servers.

How does nRouter pricing compare to OpenRouter?

nRouter charges 0% token markup — you pay exact upstream model rates — plus a flat 0% platform fee on Pay as you go credits. OpenRouter publicly lists a 5.5% fee on credit purchases (as of October 2026). Both gateways pass through underlying model costs; the difference is the platform fee on top.