Cost Analytics - MetrixLLM

The Cost Analytics page gives you a breakdown of your AI spend over time, so you can understand where costs are coming from and take action to reduce them.

Overview Chart

The top of the page shows your total spend over the selected time period as a line or bar chart. You can toggle between daily, weekly, and monthly views.

Breakdown by Dimension

Use the grouping controls to break down costs by:

Provider — Compare spend between OpenAI, Anthropic, Google, etc.
Model — See exactly how much gpt-4o vs gpt-4o-mini is costing you
API Key — Attribute spend to specific applications or teams
Workspace — See total spend if you’re an Organization admin

Cache Savings

At the top of the page, you’ll see a Cache Savings card showing how much money and how many tokens were saved by the Semantic Cache serving requests instead of hitting the upstream provider.

Filtering

The same date range and filter controls from Request Logs apply here, letting you view cost analytics for a specific model, time range, or API key.