Pricing

Pay only for what you route. No seat math. No model markups.

One rate across every provider. Subscribe to lock in a discount, or stay metered. Self-hosted is free, forever.

Self-hosted is free

The proxy is Apache-2.0. You only pay for routing through BitRouter Cloud.

01 — OVERVIEW

Three ways to pay. Same proxy, same models, same routing.

02 — PAY-AS-YOU-GO

Metered, blended, no surprises.

Blended rate
$0.50/ M tokens

Frontier models price higher, smaller models lower — the dashboard breaks it down per-model.

Includes
  • All models, all providers — one rate
  • Sub-10ms routing overhead
  • Per-key spend caps
  • Pay only for what you route
03 — SUBSCRIPTION

Lock in a lower effective rate. Overage at PAYG.

Coming soon
SOLO
$20/mo
50M tokens
$0.40 / M effective
  • 1 seat
  • 30-day usage retention
  • Community support
RecommendedTEAM
$100/mo
300M tokens
$0.33 / M effective
  • Up to 5 seats
  • 90-day usage retention
  • Priority email support
  • Per-key analytics & alerts
FLEET
$200/mo
700M tokens
$0.28 / M effective
  • Up to 20 seats
  • 12-month usage retention
  • P2P inbound routing
  • Custom routing policies

Overage billed at $0.50/M. No service interruption — set a hard cap in the dashboard if you want one.

04 — ENTERPRISE

Volume, residency, dedicated infra.

Custom plan

For teams with compliance requirements, regulated workloads, or agent fleets that can't share a control plane.

  • +Dedicated infrastructure & custom SLAs
  • +SSO, RBAC, audit logs
  • +Zero retention & in-region routing
Talk to us

Use cases, security questionnaire, deployment shapes, and contact form on the full enterprise page.

contact form · cal.com · < 24h reply

05 — ESTIMATE

Drag the slider to your monthly token volume. We'll highlight the cheapest plan in real time.

Monthly volume
50Mtok / mo
1M100M1B
  • Pay-as-you-gocheapest
    $25.0/mo
    $0.500/M effective
  • SOLOsoon
    $20.0/mo
    $0.400/M effective
  • TEAMsoon
    $100/mo
    $2.000/M effective
  • FLEETsoon
    $200/mo
    $4.000/M effective
06 — FAQ

The questions we get most often.

What counts as a token?

Both input and output tokens are billed at the same rate. Tool-call tokens, system prompts, and cached prompt tokens (where supported by the underlying provider) are passed through at provider cost.

What happens when I exceed a subscription quota?

Overage is billed at the pay-as-you-go rate, prorated to the millionth of a token. There is no hard cap or service interruption — set a spend limit in the dashboard if you want to enforce one.

Are model prices flat across providers?

No. The headline rate is a blended estimate. Frontier models (Opus, GPT-4o) cost more per token than smaller models (Haiku, gpt-4o-mini). The dashboard shows per-model rates and a usage breakdown.

Do P2P routes count against my quota?

No. Peer-to-peer routes settle directly between nodes via MPP / x402 — BitRouter does not take a cut. Subscription quotas only cover traffic routed through built-in providers.

Can I downgrade or cancel?

Yes — anytime, with no minimum term. Downgrades take effect at the next billing cycle. Cancellation stops new charges immediately and converts the account to pay-as-you-go.

Is self-hosted free?

Yes. The proxy binary is open source under the Apache-2.0 license. You only pay for routing through BitRouter Cloud, which adds telemetry, the model registry, and managed P2P discovery.