← Documentation

Savings Dashboard

Track cost savings, usage patterns, and optimization performance.

Key Metrics

Total Savings

Cumulative dollar amount saved across all optimizations

Cache Hit Rate

Percentage of requests served from cache

Tokens Saved

Total tokens reduced through compression

Total Requests

Number of API requests processed

Understanding Savings

Total Savings

The sum of savings from all optimization methods:

  • Cache savings — Cost of API calls served from cache
  • Compression savings — Cost reduction from fewer tokens
  • Routing savings — Difference between expensive and routed model costs

Cache Hit Rate

Indicates how often requests are served from cache:

80%+Excellent — Maximum cache value
50-80%Good — Normal for varied workloads
<50%Consider enabling semantic matching

Savings Breakdown

The pie chart shows contribution from each optimization method, helping identify which features provide the most value.

Time Periods

View savings for different time periods using the selector:

TodayCurrent day activity
This Week7-day rolling window
This MonthCurrent month to date
Custom RangeSelect specific dates

Savings by Model

View savings breakdown by AI model to understand which models benefit most from optimization.

ModelSaved
GPT-4$187.50
Claude-3-opus$45.20
GPT-3.5-turbo$15.13

Expensive models show higher absolute savings from caching.

Exporting Data

Export savings data for reporting or analysis.

CSVSpreadsheet format for Excel or Google Sheets
PDFFormatted report with charts
  1. Navigate to the Savings tab
  2. Select the desired time period
  3. Click Export
  4. Choose CSV or PDF format

Accessing the Dashboard

  1. Navigate to your organization
  2. Click the Savings tab

The savings dashboard is visible to all roles (Admin, Member, Viewer).

Next Steps