Cost per Thought (LLM Only)
Definition
Measures the self-contained LLM inference, excluding infrastructure cost, of one unit of AI work represented by an AI request within the observability platform. Use SDK cost estimates from the token counts x model pricing. Best for comparing efficiency between applications when used by Engineering, Product, and FinOps Practitioner Personas .
Formula, as published
Cost per Thought = LLM Estimated Cost from SDK / Total Traces
- Where:
- A “trace” is a record that captures a complete timeline of an agent’s behavior, including timestamps, errors, inputs/outputs, and contextual metadata.
- Data sources:
- Cloud Cost Billing Data Usage Telemetry (eg. Azure Cognitive Services metrics) LLM Observability/Application Telemetry
Calculate it
Example values to start with; replace them with yours.
From the FinOps Foundation
More Budgeting KPIs
Definition and formula: Cost per Thought (LLM Only), FinOps KPI library, FinOps Foundation, licensed CC BY 4.0, read 16 Oct 2026. Changed: HTML removed, dashes and British spellings changed, formula split from its data sources and examples. The calculator and the notes are Finitizer’s.
Measured once, a KPI is a number. Tracked daily, it is a practice.
Finitizer computes the KPIs your billing data supports every day, by team and product, across AWS and Google Cloud.
See how Finitizer works