Automation
Graph spend and token metrics in Grafana Cloud over OTLP, alongside the rest of your telemetry.
Your AI-coding spend lands in Grafana Cloud as Prometheus metrics, so cost sits on the same dashboards as the rest of your engineering telemetry. Lineman pushes spend, savings and token counts over OTLP every 15 minutes, with per-developer and per-repository breakdowns. Grafana Cloud rewrites the names on ingest, so you query lineman_spend_USD_total rather than lineman.spend, and you build your panels against the converted names.
Add spend and token panels to the Grafana dashboards your team already opens. lineman_spend_USD_total sits on the same time axis as the rest of your telemetry, so a cost jump and the change that caused it line up. Amounts are in US dollars.
Break the company total down with lineman_spend_by_developer_USD_total and lineman_spend_by_repo_USD_total. Developer series are labelled with the developer's email. Both breakdowns are capped at the 100 highest-spending developers and the 100 highest-spending repositories, which keeps a cumulative export inside the destination's payload limit.
Grafana Cloud stores metrics in Prometheus, so Lineman exports cumulative counters and the series query like anything else in your stack. The company token series carries a token.type label of input, output or cache. The per-developer and per-repo token series cover input and output only, because they come from the same breakdown as the rest of your Lineman dashboard.
The documentation walks through connecting it, what Lineman reads and writes, and the settings worth knowing about.
Read the setup guideEvery integration in one place, searchable by name and filterable by what it does.
View all integrations