Model usage and routing visibility.
Model usage pages explain how often models are called, how many tokens they consume, and which routes or chains carry the workload. The page is about visibility before optimization.
The page tells teams where the budget will go before it shows the bill.
Track which models are used and where the requests land.
Usage pages help teams see consumption before it turns into spend.
How often the model is called across the fleet.
How much input and output the fleet consumes.
Which model or chain handled a task.
Whether consumption is climbing or flattening.
Make usage legible enough to guide optimization.
The page makes it obvious where the product spends compute and where it saves it.
Agents or tasks that are consuming more than expected.
Different model choices shown against the same workload.
Usage as the first step toward spend control.
A basis for budgets, alerts, and model-chain decisions.