The Apache Flink Community is excited to announce the first preview release of Apache Flink Agents (0.1.0).
What is Apache Flink Agents # Apache Flink Agents is a new sub-project of Apache Flink for building event-driven AI agents directly on Flink’s streaming runtime. It unifies stream processing and autonomous agents in one framework, combining Flink’s proven strengths – scale, low latency, fault tolerance, and state management – with agent capabilities – LLMs, tools, memory, and dynamic orchestration.

Flink is an open-source stream processing framework for building real-time data applications, offering powerful features for data processing, analytics, and event-driven architectures. Developers can learn about stream processing concepts, event-driven architectures, and real-time analytics techniques to build scalable and fault-tolerant data pipelines for processing large volumes of data in real-time.

Flink

Apache Flink Agents 0.1.0 is now available as a preview release, introducing a framework for building event-driven AI agents on Flink's streaming runtime. The sub-project combines Flink's distributed processing, fault tolerance, and state management with AI agent capabilities like LLMs, tools, and memory. Key features include massive scale processing with millisecond latency, exactly-once action consistency, seamless integration with Flink's DataStream and Table APIs, and native support for both Python and Java. The release targets production use cases in e-commerce, finance, IoT, and logistics where autonomous agents must respond instantly to live events. As a preview version, it may contain issues and APIs are subject to change.

Apache Flink Agents 0.1.0 Release Announcement