跳到主要内容

Cost Tracking

CubePlex tracks LLM token usage across your organization so you can monitor spending and identify cost trends. The cost tracking dashboard is available at Admin > Cost (/admin/cost).

What is tracked​

Every LLM call made through CubePlex records:

  • Input tokens — the number of tokens sent to the model (prompt, system instructions, tool results, etc.).
  • Output tokens — the number of tokens the model generated in its response.
  • Cache read / cache write tokens — tokens served from or written to the prompt cache, billed at their own rates (often much cheaper than fresh input tokens).
  • Provider and model — which provider and model handled the call.
  • Timestamps — when the call started and ended.

Each call also records the per-model price rates (input, output, cache read, and cache write) as a snapshot at the time of the call, so historical costs stay accurate even after you later change a model's rates in Model Management.

Cost tracking also records sandbox compute events, so the totals reflect more than just LLM token spend.

Usage dashboard​

The dashboard gives you a high-level view of your organization's AI spending.

Spend over time​

A time-series chart of spend, viewable at daily or weekly granularity. You can break the series down by workspace, model, or user. Use this to spot trends — increasing usage after onboarding new team members, spikes from large batch tasks, or the impact of switching to a cheaper model.

Breakdowns​

Alongside the org-wide totals, the dashboard breaks spend down several ways for the selected period:

  • By model — which models drive the most spend, and whether cheaper alternatives might work for certain use cases.
  • By workspace — which teams or projects are spending the most.
  • By user — per-person usage.
  • By day — daily totals across the period.

Each row shows input/output and cache token counts, the call count, and the cost.

Export​

You can export the raw cost data as CSV — for the whole organization or for a single workspace — to analyze it in a spreadsheet or feed it into your own reporting.

📸 Screenshot placeholder

Capture: The Admin > Cost dashboard showing the spend-over-time chart at the top (with the workspace/model/user dimension toggle) and a breakdown table below it. Asset: /img/admin/cost-dashboard.png

Tips for managing costs​

  • Set accurate cost rates. Make sure the input and output token rates in Model Management reflect your actual provider pricing. Without accurate rates, the dashboard numbers will be misleading.
  • Review regularly. Check the dashboard weekly to catch unexpected spikes early.
  • Right-size your models. If a significant portion of spend comes from a high-cost model being used for simple tasks, consider guiding your team toward a lighter model for those use cases.
  • Use the per-model view to compare. After switching models or adding a new one, the per-model breakdown shows whether the change had the expected cost impact.