PANews reported on October 28th that Gradient, a distributed AI lab, announced the open-source release of Parallax, an operating system for local AI applications. This system supports cross-platform and cross-region deployment of large, open-source models on heterogeneous devices, including Mac and Windows, giving users complete control over models, data, and AI memory. Parallax has built-in network-aware sharding and dynamic task routing mechanisms, which can achieve intelligent scheduling based on inference load and seamlessly switch between stand-alone, multi-device, and wide-area cluster modes. Currently, Parallax is compatible with over 40 open-source models, including Qwen3, Kimi K2, DeepSeek R1, and gpt-oss. Developers can deploy locally to build and run a variety of AI applications, including programming assistants, personal agents, and multimodal generation, in a fully autonomous manner, while retaining all sensitive data and control permissions locally.PANews reported on October 28th that Gradient, a distributed AI lab, announced the open-source release of Parallax, an operating system for local AI applications. This system supports cross-platform and cross-region deployment of large, open-source models on heterogeneous devices, including Mac and Windows, giving users complete control over models, data, and AI memory. Parallax has built-in network-aware sharding and dynamic task routing mechanisms, which can achieve intelligent scheduling based on inference load and seamlessly switch between stand-alone, multi-device, and wide-area cluster modes. Currently, Parallax is compatible with over 40 open-source models, including Qwen3, Kimi K2, DeepSeek R1, and gpt-oss. Developers can deploy locally to build and run a variety of AI applications, including programming assistants, personal agents, and multimodal generation, in a fully autonomous manner, while retaining all sensitive data and control permissions locally.