Getting Started

Analytics Overview

Quick Access: Analytics | Request Logs | Cost Insights

LangMart provides comprehensive analytics to help you understand your API usage, costs, and performance. The analytics dashboard gives you visibility into every aspect of your LLM interactions.

Available Analytics

Request Logs

View detailed logs of every API request made through LangMart:

  • Request and response details
  • Token usage and costs
  • Latency metrics
  • Resolution journey (for fallback/retry scenarios)

Learn more about Request Logs

Usage Analytics

Track your usage patterns over time:

  • Requests by model and provider
  • Token consumption trends
  • Cost analysis and projections
  • Peak usage patterns

Learn more about Usage Analytics

Error Tracking

Monitor and debug failed requests:

  • Error rates and trends
  • Error breakdown by type
  • Suggested fixes
  • Historical error analysis

Learn more about Error Tracking

Dashboard Navigation

Accessing Analytics

  1. Log in to the LangMart dashboard
  2. Navigate to the Analytics section in the sidebar
  3. Choose the specific analytics view you need

Time Range Selection

Most analytics views support customizable time ranges:

  • Last 24 Hours: For real-time monitoring
  • Last 7 Days: For weekly trends
  • Last 30 Days: For monthly analysis

Filtering and Grouping

Analytics can be filtered by:

  • Model: Filter by specific model (e.g., gpt-4o, claude-3-opus)
  • Provider: Filter by provider (e.g., OpenAI, Anthropic, Groq)
  • Status: Filter by success or error status
  • Date Range: Custom date range selection

Data Granularity

Time-series data supports multiple granularity levels:

Granularity Best For Available In
Hourly Last 24 hours Short-term analysis
Daily 7-30 days Trend identification
Weekly 30+ days Long-term patterns
Monthly 90+ days Quarterly reviews

Key Metrics

Cost Metrics

  • Total Cost: Sum of all request costs
  • Cost by Model: Breakdown by model
  • Cost Projections: Estimated future costs based on trends

Performance Metrics

  • Latency: Time from request to response
  • Time to First Token (TTFT): Time until streaming begins
  • Success Rate: Percentage of successful requests

Token Metrics

  • Input Tokens: Tokens sent to the model
  • Output Tokens: Tokens received from the model
  • Total Tokens: Combined token usage

Setting Up Alerts

Combine analytics with alerts to get proactive notifications:

  1. Navigate to Alerts in the sidebar
  2. Create alerts for:
    • Cost thresholds
    • Error rate spikes
    • Usage anomalies
    • Quota warnings

See the Alerts documentation for detailed setup instructions.

Exporting Data

Analytics data can be accessed programmatically via the API:

# Get usage analytics
curl -X GET "https://api.langmart.ai/api/account/analytics/usage?start_date=2025-01-01&end_date=2025-01-31" \
  -H "Authorization: Bearer YOUR_API_KEY"

# Get request logs
curl -X GET "https://api.langmart.ai/api/account/request-logs?limit=100" \
  -H "Authorization: Bearer YOUR_API_KEY"

# Get error analytics
curl -X GET "https://api.langmart.ai/api/account/errors?start_date=2025-01-01" \
  -H "Authorization: Bearer YOUR_API_KEY"
Feature Direct Link
Analytics Dashboard https://langmart.ai/analytics
Request Logs https://langmart.ai/requests
Usage https://langmart.ai/usage
Cost Insights https://langmart.ai/analytics/cost-insights
Errors https://langmart.ai/errors
Alerts https://langmart.ai/alerts

Next Steps