Usage
Track token consumption, latency, and execution volume across your project.
Compass records usage at every level — from a single step up to the whole project — so you can see how much your AI is doing and how efficiently. It's all surfaced through Observability.
Token usage
Token counts are tracked everywhere a model is involved:
- Per step — each agent step and tool call in a trace shows the tokens it used.
- Per execution — the execution's header totals the tokens across all its steps.
- Per agent or workflow — the rollup table totals tokens across all of an entity's executions.
- Project-wide — the metrics panel shows total tokens over the last 30 days.
This lets you find which work is token-heavy, from the project level all the way down to the individual step.
Latency
Duration is tracked the same way — per step, per execution, and aggregated per agent or workflow as an average and a p95 (the 95th-percentile duration, a good measure of worst-case slowness). The project metrics panel shows overall p95 and average latency.
Execution volume and success
Counts and success rates roll up across executions:
- executions over the last 30 days, project-wide,
- executions over the last 24 hours, per deployment,
- and success rate at the project, per-entity, and per-deployment levels.
Together these tell you how busy each piece of work is and how reliably it's completing.
Cost
Token usage is priced into a dollar figure using each model's rates, and cost rolls up the same way tokens do — per step, per execution, per agent or workflow, and project-wide. The Spend tab charts spend over time and your top-spending agents and workflows.
For a model without published pricing (a local model, for example), set per-token rates on its provider connection to enable cost tracking for everything that runs through it.
To put a budget on spend — and block new work when it's hit — see Spend limits.