MLflow

MLflow is an open-source platform for managing the complete machine learning lifecycle, including experimentation, reproducibility, deployment, and a central model registry. MLflow supports both traditional ML and generative AI workflows:

- MLflow Tracking: Record and query experiments, including code, data, config, and results. Now with integrated tracing for GenAI workflows across multiple frameworks.
- MLflow Models: Deploy machine learning models in diverse serving environments, with GenAI support for ChatModels and streaming interfaces.
- Model Registry: Store, annotate, discover, and manage models in a central repository.
- MLflow Evaluation: Evaluate model performance using customizable metrics, including LLM-as-judge frameworks and GenAI-specific benchmarks.
- MLflow Deployments: Simplify model deployment and serving across various platforms, with expanded capabilities for hosting large language models.