Back to all tools
Evidently AI
Open-source ML/LLM monitoring — data drift reports and quality test suites.
0Open Source
AI Observability
Overview
Evidently is an open-source Python library for evaluating, testing, and monitoring ML and LLM models — generating interactive reports and test suites for data drift and model quality.
Key Features
- Data drift detection across tabular, text, and embeddings
- LLM evaluation metrics: faithfulness, relevance, toxicity
- Interactive HTML reports for sharing with stakeholders
- Test suites for automated data quality CI checks
- Evidently Cloud for production monitoring dashboards
- Supports sklearn, pandas, and any ML framework
Real-World Workflows
Add ML model monitoring to CI/CD
- 1Load reference (training) and current (production) data
- 2Create a TestSuite with data drift and quality tests
- 3Run in CI pipeline — test suite fails if drift exceeds threshold
- 4Generate HTML report artifact for stakeholder review
Getting Started
pip install evidently
from evidently.report import Report
from evidently.metric_preset import DataDriftPreset
report = Report(metrics=[DataDriftPreset()])
report.run(reference_data=ref_df, current_data=cur_df)
report.save_html('drift_report.html')Compare Alternatives
See how Evidently AI stacks up against similar tools.