Usage

Spend, savings, tokens and latency for your organization.

📄️Get usage overview

Headline numbers for the period: requests, spend against the baseline, tokens, latency and routing efficiency. `period` is `24h`, `7d`, `30d` or `90d`.

📄️Get usage time series

Requests, spend, tokens and latency per time bucket. `granularity` (`hour`, `day` or `week`) defaults to a sensible value for the period.

📄️Get usage breakdown

Usage for the period, split by provider, by model and by routing strategy.

📄️Get recommendations

Suggested routing changes based on your traffic in the period, with estimated savings.