Cleanlab is an open-source Python library that quickly identifies dataset problems in machine learning projects, offering a data-centric AI platform to run algorithms and detect issues such as mislabeling, outliers, near duplicates, drift, etc. The latest release of cleanlab v2.6 greatly expands its capabilities, including comprehensive issue detection in Datalab, automatic flagging of null values, alerting for imbalanced classes, discovery of underperforming groups, data valuation, and support for multiple ML tasks, including object detection. Additionally, the library has been enhanced with better scaling, efficient label issue detection, and improved performance in binary classification tasks. The cleanlab community continues to grow with new contributors, and the project aims to empower data scientists and researchers with a free and transparent tool to improve dataset quality for reliable machine learning.