mlflow 3.1.3

v3.1.3

on Python PyPI

MLflow 3.1.3 includes several features and improvements

Features:

• [Artifacts / Tracking] Do not copy file permissions when logging artifacts to local artifact repo (#16642, @connortann)
• [Tracking] Add support for OpenAI ChatCompletions parse method (#16493, @harupy)

Bug fixes:

• [Deployments] Propagate MLFLOW_DEPLOYMENT_PREDICT_TIMEOUT to databricks-sdk (#16783, @bbqiu)
• [Model Registry] Fix issue with search_registered_models with Databricks UC backend not supporting filter_string (#16766, @BenWilson2)
• [Evaluation] Bug fix: Databricks GenAI evaluation dataset source returns string, instead of DatasetSource instance (#16712, @dbczumar)
• [Tracking] Fix the position of added tracking_uri param to artifact store implementations (#16653, @BenWilson2)

Small bug fixes and documentation updates:

#16786, #16692, @daniellok-db; #16594, @ngoduykhanh; #16475, @harupy