NVIDIA and Microsoft are accelerating the evolution of PC software by integrating powerful generative Artificial Intelligence capabilities into Windows 11 through NVIDIA RTX AI PCs. Central to this progression is the reimagined NVIDIA TensorRT, now optimized for RTX AI PCs with just-in-time, on-device engine building and a significantly reduced package size, enabling faster and more efficient deployment of AI-powered features to over 100 million devices.
Unveiled at Microsoft Build, TensorRT for RTX is now natively supported by Windows ML, a new inference stack that offers developers both broad hardware compatibility and enhanced performance. Windows ML leverages ONNX Runtime and can automatically select the optimal compute resource, from GPU to CPU or NPU, for each AI task. This integration delivers over 50% faster performance than DirectML for AI workloads on GeForce RTX GPUs, removing the need for developers to manage hardware-specific execution files within their apps. Developers immediately gain access to the latest TensorRT performance improvements, streamlining deployment and maintenance processes.
NVIDIA provides a comprehensive suite of development tools and SDKs tailored for AI and creative workflows. These include CUDA and TensorRT for hardware acceleration, DLSS and Optix for 3D rendering, RTX Video and Maxine for multimedia, and Riva and ACE for generative AI tasks. Leading software vendors like Autodesk, Bilibili, Chaos, LM Studio, and Topaz Labs are already releasing updates to tap into RTX AI features and enhanced app performance. Notably, NVIDIA´s NIM microservices offer developers prepackaged, optimized AI models for seamless local deployment and accelerated performance, including new support for image generation models and expanded compatibility with the latest GeForce RTX GPUs.
NVIDIA also spotlights Project G-Assist, an experimental AI assistant integrated into the NVIDIA app, empowering users to control their PCs, peripherals, and gaming environments via natural language. Developers can quickly build and distribute plug-ins using the no-code Plug-in Builder, with recent community additions supporting services like Google Gemini, Spotify, Twitch, IFTTT, and SignalRGB. The AI ecosystem continues to expand—with G-Assist integration into platforms like Langflow, enabling low-code workflows for more flexible AI applications. NVIDIA´s tools and initiatives, built in partnership with Microsoft, aim to lower barriers for AI development on Windows 11 and foster a thriving ecosystem of productivity, creativity, and intelligent automation at scale.
