Skip to main content
Docs
Analytics

Analytics

Track usage, costs, and performance across your organization's AI operations.

The Analytics dashboard provides visibility into your AI usage, costs, and performance metrics. Monitor trends, identify optimization opportunities, and control spending.

Accessing Analytics

  1. Sign in to studio.elyxir.ai
  2. Go to Analytics in the main navigation
  3. View dashboards and reports

Dashboard Overview

The main dashboard shows:

  • Total Requests: API calls over time
  • Token Usage: Input and output tokens
  • Costs: Spending by model and time
  • Performance: Latency and error rates

Time Range Selection

Filter data by time period:

RangeDescription
24 hoursLast day
7 daysLast week
30 daysLast month
90 daysLast quarter
CustomSelect dates

Key Metrics

Request Metrics

MetricDescription
Total RequestsNumber of API calls
SuccessfulRequests that completed
FailedRequests that errored
Rate LimitedRequests that hit limits

Token Metrics

MetricDescription
Input TokensTokens in prompts
Output TokensTokens in responses
Total TokensCombined usage
Average per RequestTokens per call

Cost Metrics

MetricDescription
Total CostOverall spending
Cost per ModelBreakdown by model
Cost per KeyBreakdown by API key
Daily AverageAverage daily spend

Filtering and Grouping

Filter By

  • Model (GPT-4, Claude, etc.)
  • API Key
  • Date range
  • Status (success/error)

Group By

  • Day / Week / Month
  • Model
  • API Key
  • Hour of day

Charts and Visualizations

Usage Over Time

Line chart showing:

  • Requests per period
  • Token usage trends
  • Cost accumulation

Model Distribution

Pie chart showing:

  • Usage by model
  • Cost by model
  • Request distribution

Performance Metrics

Gauge and bar charts for:

  • Average latency
  • Success rate
  • Error distribution

Exporting Data

Export analytics for reporting:

  1. Configure your filters
  2. Click Export
  3. Choose format (CSV, Excel)
  4. Download report
Analytics | elyxir