Introducing Apache Airflow® 3: the most significant release in Airflow's history

Apache Airflow 3 marks a pivotal upgrade in the evolution of the open-source data orchestration tool, incorporating significant architectural and functional improvements tailored for the AI era. Trusted by over 77,000 organizations, this release introduces features like DAG Versioning, a modern UI, stronger security with task isolation, and the ability to run tasks remotely and at any time, enhancing usability and operational agility. The new distributed architecture decouples task execution from direct metadata database connections, supporting more secure and scalable workflows. With the rising importance of generative AI and complex data pipelines, Airflow 3 is positioned to meet these evolving demands, allowing for event-driven scheduling and inference execution. The release underscores the community-driven nature of Airflow, with contributions shaping its new functionalities, and positions Airflow as a robust tool for managing business-critical applications and revenue-generating use cases. As organizations increasingly rely on data for competitive advantage, Airflow 3 is designed to support sophisticated and large-scale data workflows, ensuring its continued relevance and utility in modern data environments.