Datadog LLM Observability

Datadog LLM Observability provides end-to-end monitoring for LLM applications within the Datadog platform. It offers trace visualization, prompt/response inspection, cost tracking, and quality evaluations alongside your existing APM data.

Datadog LLM Observability is a monitoring solution integrated within the broader Datadog observability platform. It provides end-to-end visibility into large language model (LLM) applications, allowing teams to trace requests, inspect prompts and responses, track costs, and evaluate output quality alongside existing application performance monitoring (APM) data. The tool is designed for developers and operations teams who build and maintain LLM-powered features, offering a unified view of both traditional application metrics and LLM-specific telemetry. As part of the Datadog ecosystem, it benefits from integrations with cloud providers, Kubernetes, and AI services, and can be used in conjunction with other Datadog products like logs, metrics, and security monitoring.