For AI agents: a documentation index is available at the root level at /llms.txt and /llms-full.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
Trust CenterStatusSupportGet a demoPlatform
DocumentationEvals API ReferenceIntegrations & OTELPlatform SettingsSelf-HostingChangelog
DocumentationEvals API ReferenceIntegrations & OTELPlatform SettingsSelf-HostingChangelog
    • Platform Settings
    • Data Residency
    • RBAC
  • Project Settings
    • API Keys
    • Team Members
    • Roles & Permissions
    • Transformers
    • Integrations
    • Alerts
    • AI Connections
    • Model Costs
    • Data Usage
    • Evaluation Models
    • Evaluation Rules
    • Annotation Options
    • Classifiers
    • Data Sources
    • Data Retention
    • Audit Logs
  • Organization Settings
    • Projects
    • Users
    • Roles & Permissions
    • Model Credentials
    • SSO
    • Data Retention
    • Audit Logs
    • Feature Access
LogoLogo
Trust CenterStatusSupportGet a demoPlatform
On this page
  • Overview
  • Usage Summary
  • Ingested Trace/Span Data
  • Online Evals Ran Per Month
  • Signals Ran Per Month
  • Cost Insights
  • Observability Data Usage
  • Online Evals Usage
  • Signals Usage
Project Settings

Data Usage

View processed data and retention usage for your project.
Was this page helpful?
Previous

Evaluation Models

Configure and manage the evaluation models used for running LLM-as-a-judge metrics in your project.

Next
Built with

Data Usage lets you monitor your project’s data consumption across traces, spans, online evaluations, and test runs. Use this page to understand your usage patterns and track progress toward your plan limits.

Data Usage Overview

Overview

The Overview tab provides a breakdown of usage by data type over a selected time period.

Usage Summary

Usage is broken down into five categories:

Data TypeDescriptionBilling
TracesTop-level observability units representing complete interactionsContributes to GB usage
SpansIndividual operations within tracesContributes to GB usage
Online Evals (metric data)Evaluations run on production traces, spans, and threadsCounted separately
Test RunsEvaluation runs from experiments and test suitesCounted separately
SignalsClassifier runs on traces and idle threadsCounted separately

The summary shows:

  • Total in period — Total data size and unit count for the selected time range
  • Average per month — Average monthly data size and unit count

Ingested Trace/Span Data

This section shows your observability data ingestion over time. Your observability data usage is calculated based on GB-months—the total gigabytes of traces and spans ingested over your selected retention period.

You can view the data in three ways:

  • Ingested size over time — Total data size ingested per month
  • Count over time — Number of traces/spans ingested per month
  • Avg size per trace/span — Average size of each trace or span

You’ll only be charged for additional data you ingest OR retain beyond your plan’s limits. Adjust your data retention settings to manage costs. The minimum retention period is 1 month.

Online Evals Ran Per Month

This section shows your online evaluation usage over time. Online evals are evaluations run on your production traces and spans in real-time. You’ll be charged based on the total number of metric evaluations run beyond your plan’s included limits.

The chart displays the number of online evals executed per month, helping you track evaluation volume and identify usage trends.

Signals Ran Per Month

This section shows your classifier usage over time. Each classifier evaluation on a trace or idle thread is counted as one signal run. You’ll be charged based on the total number of signals run beyond your plan’s included allowance.

The chart displays the number of signals executed per month, scoped to whichever sample rates you’ve configured for trace and thread classifiers in Project Settings → Classifiers.

Cost Insights

The Cost Insights tab shows how close you are to your plan’s included usage limits.

How usage contributes to costs depends on your plan:

  • Free, Starter, and Premium plans — Usage limits are set at the project level. Each project has its own usage allowance, and the data shown reflects how much this project contributes to its own costs.
  • Team and Enterprise plans — Usage limits are set at the organization level. All projects share a pooled allowance, and the data shown reflects how much this project contributes to your organization’s overall costs.

Billing is managed at the organization level for all plans—the difference is whether usage limits are tracked per project or pooled across the organization.

Observability Data Usage

Tracks your trace and span data consumption against your plan’s GB-months allowance. The progress bar shows your current usage as a percentage of your included limit, with a breakdown between traces and spans.

GB-months is calculated by summing your trace/span data ingested over your selected retention period. You can reduce usage by shortening your retention period.

Online Evals Usage

Tracks the number of online evaluations run on your production traces, spans, and threads in real-time. Online evals are evaluations that run automatically on incoming observability data, as opposed to test runs which are triggered manually or via CI/CD.

The progress bar shows your current eval count against your plan’s included allowance.

Signals Usage

Tracks the number of classifier runs against your project’s traces and threads. Each enabled classifier that processes an item counts as one signal run, subject to the configured trace and thread sample rates.

The progress bar shows your current signal count against your plan’s included allowance. To control overhead, lower the sample rate on individual classifiers in Project Settings → Classifiers, or disable classifiers you no longer need.