Splunk AI Token and Cost Monitoring

Monitor, control, and alert on your organization's LLM token usage and AI resource costs with Splunk AI Token and Cost Monitoring.

Monitor, control, and alert on your organization's LLM token usage and AI resource costs with Splunk AI Token and Cost Monitoring. Get comprehensive visibility into the token costs associated with using and deploying AI, from the infrastructure powering your AI applications to the first-party AI agents integrated with your organization.

You can use the Splunk AI Token and Cost Monitoring dashboards to:

  • Track application and provider-driven token costs.

  • Attribute application costs to agents, providers, and models.

  • Investigate sources of changes in costs and token usage, such as agents, providers, models, and tool calls.

  • Track infrastructure cost by different providers and models.

Monitor AI agent token usage and costs

To use the AI agent tokens and costs dashboard, you must Collect traces and metrics from AI agents and applications.

To access the dashboard, use the Splunk Observability Cloud main menu to select APM > AI agent tokens and costs.

You can use the dashboard to track:

  • The number and cost of total tokens, input tokens, and output tokens used by first-party AI agents.

  • The number and cost of input vs. output tokens used by your AI agents over time.

  • The number, percentage, and cost of tokens used by individual AI agents, models, operations, and providers over time.

Filter the data on the dashboard by agent, provider, or model using the filters above the charts.

Monitor AI infrastructure token usage and costs

To use the AI infrastructure provider tokens and costs dashboard, set up one or more of the following data integrations:

To access the dashboard, use the Splunk Observability Cloud main menu to select Infrastructure > AI infrastructure provider tokens and costs.

You can use the dashboard to track:

  • The total number and cost of total tokens, input tokens, and output tokens used by your AI infrastructure components.

  • The number and cost of input vs. output tokens used by your AI agents over time.

  • The number, percentage, and cost of tokens used by individual AI agents and models.

Filter the data on the dashboard by a specific time period using the filter above the charts.

To edit token rates, select the actions (…) menu in the upper-right corner of the dashboard. In the Edit token rates panel, you can add and remove models or modify the input, output, and cached token rates.