京东开源 | JDOpenSource

Most existing open-source agent projects are primarily SDKs or frameworks, requiring users to perform additional development and lacking true out-of-the-box usability. In contrast, our open-source JoyAgent-JDGenie is an end-to-end multi-agent product that can directly answer or resolve user queries or tasks. For example, when a user submits a query like"Provide an analysis of recent trends between the US dollar and gold,"JoyAgent-JDGenie can instantly generate a report in web or PPT format.

xLLM is an efficient LLM inference framework, specifically optimized for Chinese AI accelerators, enabling enterprise-grade deployment with enhanced efficiency and reduced cost. The framework adopts a service-engine decoupled inference architecture, achieving breakthrough efficiency through several technologies: at the service layer, including elastic scheduling of online/offline requests, dynamic PD disaggregation, a hybrid EPD mechanism for multimodal and high-availability fault tolerance; and at the engine layer, combined with technologies such as multi-stream parallel computing, graph fusion optimization, speculative inference, dynamic load balancing and global KV cache management.

With the widespread adoption of generative AI applications, risks such as content safety violations, data leakage, and prompt injection attacks are becoming increasingly prominent. Ensuring that large model outputs comply with regulatory requirements and avoiding security vulnerabilities have become critical challenges that enterprises must address.To tackle these issues, JD.com is officially open-sourcing its large-model safety framework—JoySafety. This framework has been extensively used within JD.com across diverse scenarios, including AI-powered shopping assistants, logistics customer service, sales assistants, medical consultations, merchant workbenches, legal advisory services, and security Q&A systems. JoySafety supports hundreds of millions of daily inference calls and achieves a >95% attack interception rate, aiming to provide enterprises with a mature, reliable, and free security solution for large language models.

JoySafeter is not just a productivity tool, but an "operating system" for security capabilities. It unifies fragmented security tools into a collaborative AI army through visual intelligent orchestration, and precipitates individual expert experience into organizational digital assets. It is the first to define a new paradigm of AI-driven Security Operations (AISecOps) in the industry.

OxyGent is an open-source framework that unifies tools, models, and agents into modular Oxy. Empowering developers with transparent, end-to-end pipelines, OxyGent makes building, running, and evolving multi-agent systems seamless and infinitely extensible.

JDOxyGent4J shares the same philosophy as the Python version and is deeply optimized for the Java ecosystem: native type safety and compile-time validation, seamless Spring Boot integration, enterprise-grade concurrency and stability, and developer-friendly APIs for Java developers.