Monitoring & Analytics
JetPero gives you real-time visibility into your API activity across all projects, providers, and endpoints β with beautiful charts and actionable insights.
π What You Can Track
β Total Requests
π Requests per Provider
π Success vs Error Rates
π¦ Payload Size & Response Time
π Retry & Timeout Stats
π Geo & IP Distribution
π Auth Issues & Blocked Calls
π Historical Usage Trends
All data is filtered per project, ensuring isolated analytics for each environment or client.
π Real-Time Dashboard
Every project has a real-time dashboard that updates automatically.
Youβll see:
β Top performing endpoints
β οΈ Error breakdown by status code (400s, 500s)
π§ Usage heatmaps (per day/hour)
π¬ Top messages (e.g., for LLM/chat apps)
π Provider call distribution
Perfect for devs, product teams, and ops who need API clarity fast.
π Historical Trends
Navigate to: Dashboard
Youβll get:
β³ Daily / weekly / monthly usage patterns
π Performance degradation tracking
π Request spikes and drops
Use filters by:
π Project
π§© Provider
π Endpoint
π Time range
π οΈ Debug Requests with Logs
Every request is logged under Request Logs, where you can:
π View request body and headers (truncated for security)
π΅οΈ See the injected provider key (hashed)
β οΈ Inspect errors and timeouts
π Replay failed requests (coming soon)
Debugging latency or failure has never been easier.
π Custom Alerts (coming soon)
Weβre working on custom alerting, so you can:
π Get notified when an error rate spikes
π Be warned if request volume drops
π¬ Slack / Email / Webhook alerts
Stay ahead of issues before your users notice them.
π§ Developer-Focused Insights
JetPero is built for devs, so our analytics is not just eye-candy:
π€ Breakdown by SDK / language
π§ͺ Experiment tracking (coming soon)
π§΅ Traces between retries & fallbacks
π₯ Rate limit vs actual traffic visualizations
π‘ Use Cases
β Optimize API usage for cost and performance
π¨ Quickly identify provider outages
π Fine-tune LLM prompts using most common queries
π Detect security issues early (bad tokens, repeated failures)
Last updated