Dynatrace AI Observability

platform

proprietary-SaaS(Grail-powered, Davis AI)

license

commercial

ingestion

OpenTelemetry+OpenLLMetry(via Traceloop)+GenAI-semconv(open standards, not open product)

source

closed

Fully proprietary Dynatrace platform; AI Observability is a commercial app/module. OTel + OpenLLMetry ingestion are open standards consumed by a closed product. Scored closed (1).

• https://docs.dynatrace.com/docs/observe/dynatrace-for-ai-observability recorded 2026-08-13

  Documentation for the capability - "Use Dynatrace with Traceloop OpenLLMetry, OpenTelemetry with GenAI semantic conventions, or OpenInference" to observe AI-powered cloud applications end-to-end. Those are ingestion formats. Nothing on the page publishes source for Dynatrace itself or offers a build of it.

Dynatrace is a large public observability vendor, but no primary usage figure specific to AI Observability is published: the documentation confirms the product exists and carries no customer count, no account count and no ingested-volume figure for it. Rather than attribute platform-wide numbers that could not be sourced primarily, no level is assigned. The absence is confirmed rather than merely unchecked.

• https://docs.dynatrace.com/docs/observe/dynatrace-for-ai-observability recorded 2026-08-13

  The capability's documentation, read for a usage figure and containing none. It describes what the product monitors - token consumption, latency, availability, errors, model drift and service fees across Amazon Bedrock, NVIDIA NIM, Ollama and others - and never how many people use it.

Strong full-stack AI observability: end-to-end prompt-flow tracing and debugging with root-cause analysis, prompt latency and model-level metrics; predictive cost management through Davis AI with token and service-fee tracking and full capture without sampling; OpenTelemetry, OpenLLMetry and the GenAI semantic conventions; and providers spanning OpenAI, Bedrock, NIM, Ollama, Gemini and Azure AI Foundry alongside LangChain and CrewAI. It lacks the dedicated evaluation, LLM-as-judge and prompt-versioning depth of the evaluation-first tools, so it sits level with Langfuse at 4.

• https://docs.dynatrace.com/docs/observe/dynatrace-for-ai-observability recorded 2026-08-13

  Documentation - trace "LLM calls, tool-use, RAG, and resolve performance, latency, cost, and reliability issues"; Cost covers "Token usage, service fees, or overall resource consumption"; drift is anticipated by "monitoring input data stationarity"; and providers named include Amazon Bedrock, NVIDIA NIM and Ollama.