Observability Suite
Track prompts, costs, and model metrics in real time.
Lightweight, high-performance tracing for LLM applications. Analyze model behavior, trace latencies, monitor tokens, and capture anomalies.
Live Chat Console
Start a new sandbox session, stream prompts to model providers, and trace logs automatically.
Configure
Run History
Browse previous conversation logs, inspect prompt structures, and review completed chat threads.
Configure
Analytics metrics
Graph request volume, p95 latency, provider token distribution, and service error rates.
Configure
Anomaly detection
Spot token spikes, backend failures, and latency regressions dynamically from query logs.
Configure
