Metering, cost and limits

Local-only mode

For gateways that only front your own GPUs, hide every price and keep metering tokens and speed.

What it is

For self-hosted deployments with no billing relationship, JANUS_LOCAL_ONLY=true turns off cost tracking and hides every price and spend surface.

Why you want it

Teams running only their own GPUs care about tokens and speed, not dollars. Local-only mode removes pricing chores and noise.

How it works

  • Usage records $0 while token, request and latency metering continue
  • Models can be enabled without a rate card
  • USD quotas cannot be created; existing ones become inert
  • Reporting redacts monetary data server-side

Related features

Metering that matches the invoice

Every request is metered by input, output and cache tokens at the provider's real prices, with no prompts stored.

Details →

Versioned rate cards

Up to five prices per model with full history, so past requests keep the price that applied when they ran.

Details →

Quotas and budgets

Cap tokens, spend or requests per person, team or service, with warnings before anyone hits the limit.

Details →

See it on your own network.

The Community edition is free for up to 25 people. The 30-day Business trial unlocks every Business feature.

Download free Start a Business trial