Library / SDK
Observability & Evaluation
Evidently
1Evaluate 2test 3and 4monitor 5ML 6models
the six Ws · specification
W1
Who
Maintained by Evidently AI.
W2
What
A framework for evaluating, testing, and monitoring ML and LLM systems, including drift and quality.
W3
Where
A Python library run locally or in CI, with an optional cloud.
W4
When
Reach for it when you need drift detection or LLM evaluation reports in production.
W5
Why
A broad library of prebuilt metrics and reports for ML and LLM systems.
W6
With
Works with pandas and scikit-learn; integrates with Grafana and Airflow.
W7
Watch
Maintained by Evidently AI; open-source (Apache-2.0) library, self-runnable. Reports compute locally; the optional cloud (open-core) receives uploaded data.