Metering, cost and limits
Versioned rate cards
Up to five prices per model with full history, so past requests keep the price that applied when they ran.
What it is
Each model has a rate card with up to five prices per million tokens: input, output, cached input, 5-minute cache write and 1-hour cache write. Past requests keep the rate in force when they ran.
Why you want it
Prices change. Versioned rates keep historical reports correct and give an audit trail for every price change.
Discovered models, their status, five-part rate cards, context windows and grants.
Actual Janus interface · Demo identities and synthetic usageHow it works
- Bundled reference prices for common OpenAI, Anthropic, Mistral and Gemini models; xAI and OpenRouter prices imported at discovery
- An explicit $0 card enables self-hosted models free of charge
- The pricing dialog explains each dimension in the provider's own terms
- Every change is audit-logged
In the product
Screenshot above. Actual Janus interface with demo identities and synthetic usage.
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.