OpenTelemetry-based tracing and observability for prompts, model calls, tool spans, tokens, latency, and errors in production AI applications.
Features
- OpenTelemetry compatibility
- Prompt and model-call tracing
- Token and cost visibility
- Tool-call spans
- OTLP data export
Use Cases
- Production LLM monitoring
- Agent trace debugging
- Latency bottleneck analysis
- Token-cost governance
- Error root-cause analysis
FAQ
OpenTelemetry-based tracing and observability for prompts, model calls, tool spans, tokens, latency, and errors in production AI applications. Core capabilities include: OpenTelemetry compatibility, Prompt and model-call tracing, Token and cost visibility.
Common scenarios include: Production LLM monitoring, Agent trace debugging, Latency bottleneck analysis.
