xAI usage
Spend by day and model, for one team. From xAI's usage report. Never a prompt or a reply.
Access
- Kind
- A read-only key you paste
- What to give
- Management key and team ID
- Why that one
- An xAI management key, which is separate from an API key. Valistry asks the team's usage endpoint for dollars per day and model, and calls nothing else.
What it attributes on its own
- From the source
- day; model; endpoint kind (chat, image); what xAI charged
- Needs a rule or a trace
- tokens — the reference names no token field, so none is asked for; API key — spend is per team; workflow; agent; outcome
The addresses it calls
- https://management-api.x.ai/v1/billing/teams/
Valistry calls only these addresses. It blocks any other call and logs every call.
Connecting it
- Create a management key. Settings (bottom left) → Management Keys → Create management key Open ↗
- Copy the team ID. Settings → Team, then copy the team ID. Open ↗
OpenRouter, a gateway or OpenTelemetry report xAI tokens; this source reports the bill. Keep this one for xAI and disconnect the other, or each call counts twice.