Observability is built-in! With Helicone Cloud, all requests are automatically logged and monitored. No setup required.View your requests, costs, and analytics instantly in your Helicone dashboard.

​
What’s automatically tracked:

  • Request logs - Every request and response with full details
  • Cost tracking - Real-time costs across all providers
  • Latency metrics - Performance insights and bottleneck detection
  • Token usage - Detailed token analytics by model and user
  • Error rates - Monitor failures and retry patterns
  • Custom properties - Tag requests with metadata for filtering

​
Privacy Controls

Even with automatic logging, you maintain full control over sensitive data:
curl https://gateway.helicone.ai/your-org-id/ai/chat/completions \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "helicone-omit-request: true" \
  -H "helicone-omit-response: true" \
  -H "Content-Type: application/json" \
  -d '{"model": "gpt-4o-mini", "messages": [{"role": "user", "content": "Sensitive data"}]}'