1. [Topics](/topics) [Artificial Intelligence](/topics/ai/) [Local AI](/local-ai/) Port Your Windows Application to RTX # Port Your Windows Application to RTX Spark Introducing the [NVIDIA RTX Spark](https://www.nvidia.com/en-us/products/rtx-spark/)™ Superchip. The fusion of NVIDIA AI and RTX graphics in a single chip redefines Windows PCs and delivers amazing creating, AI development, and gaming. Port your app to this new Windows-on-Arm platform now to deliver a fast, reliable native experience your users expect and join the [100+ software providers](https://nvidianews.nvidia.com/news/nvidia-microsoft-windows-pcs-agents-rtx-spark) already embracing the platform. [Get Started ](#get-started)[Access Porting Guide](https://docs.nvidia.com/rtx/rtx-spark-porting-guide/index.html) * * * ## Windows on Arm Resources ### What is Windows on ARM? Windows on Arm is a version of the Windows operating system that offers high-performance and energy efficient computing on NVIDIA RTX Spark PCs. [Learn About Windows on ARM ](https://www.arm.com/glossary/windows-on-arm) ### Windows on Arm Developer Overview Explore tools and guidance to build, port, test, and optimize Windows apps for native Arm64 performance on NVIDIA RTX Spark PCs. [Port your Windows App to Arm](https://learn.microsoft.com/en-us/windows/arm/add-arm-support) [Arm64EC Guide](https://learn.microsoft.com/en-us/windows/arm/arm64ec) [See all ARM Resources](https://learn.microsoft.com/en-us/windows/arm/overview) ### Windows on ARM FAQs Find answers to common questions about Windows on Arm PCs, including app compatibility, performance, and device support. [Read Windows on ARM FAQs](https://support.microsoft.com/en-us/windows/experience/platform-variants/windows-arm-based-pcs-faq) * * * [![](https://developer.download.nvidia.com/images/local-ai/Windows_On_ARM.png)](https://developer.download.nvidia.com/images/local-ai/Windows_On_ARM.png) _Click diagram to enlarge_ ## Windows-on-Arm Migration Options Windows on Arm offers three deployment paths: - **Emulation** : Run existing x86 software without modification. - **ARM64EC** : Port performance-critical code while retaining legacy x86 dependencies. - **Full ARM64** : Deliver fully native execution with the greatest engineering effort. Start with emulation, then port components where performance or platform requirements justify it. * * * ## Steps to Port Your Application You can begin porting your application before RTX Spark hardware is available. Use an existing Windows on Arm development system to identify incompatible dependencies and produce a Arm64 build. 1. Review your application and dependencies for Arm64 support. 2. Choose an Arm64 or Arm64EC porting strategy appropriate for your application. 3. Build and test your application on Windows on Arm. 4. Validate the NVIDIA and CUDA software paths used by your application. 5. Test installation, updates, functionality, and performance. 6. Validate on RTX Spark once supported hardware and software become available. * * * ## Get Started Porting Your Application Access the following resources to port your application ### NVIDIA x64 Driver for Cross Compilation Developer driver to cross-compile for Windows-on-Arm devices. [Download](https://www.nvidia.com/en-us/drivers/details/272764/) ### CUDA Toolkit 13.4 CUDA Toolkit 13.4 enables native Windows Arm64 development and cross-compilation from the Windows x86\_64 toolkit. [Download](https://developer.nvidia.com/cuda-downloads)[Documentation and Release Notes](https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html) ### Nsight Tools NVIDIA Nsight Graphics and NVIDIA Nsight Aftermath SDK support graphics debugging, profiling, and GPU-crash diagnostics. [Download Nsight Graphics](/nsight-graphics/get-started)[Download Nsight Aftermath](/nsight-aftermath/getting-started) ### TensorRT for RTX TensorRT for RTX brings optimized AI inference and cutting-edge acceleration to developers using NVIDIA RTX GPUs. [Download TensorRT for RTX](/tensorrt-rtx)[TensorRT RTX Execution Provider 0.4.0](https://github.com/NVIDIA/TensorRT-RTX-EP-ABI/releases/tag/v0.4.0) ### Ollama on Windows ARM64 Ollama is an easy-to-use LLM inference tool that runs models based on the llama.cpp engine. [Download](https://github.com/ollama/ollama/releases/tag/v0.32.3) ### PyTorch on Windows Arm64 PyTorch is a foundational framework for AI that is built for diffusion, fine-tuning and experimentation. Download using: python -m pip install --pre --upgrade --only-binary=:all: --extra-index-url 'https://pypi.nvidia.com/nvtorch\_oot\_nightly' torch torchvision torchaudio [View Instructions](https://pypi.nvidia.com/nvtorch_oot_nightly/) ### Llama.cpp on Windows ARM64 Llama.cpp is an light-weight C/C++ framework for LLM inferencing. [Download Llama.cpp ARM64](https://github.com/ggml-org/llama.cpp/releases/tag/b10360) * * * ## More Resources ![Sign up for the NVIDIA Developer Newsletter](https://developer.download.nvidia.com/icons/m48-email-settings.svg) ### Sign up for the Developer Newsletter ![Join the Developer Community](https://developer.download.nvidia.com/images/omniverse/m48-people-group.svg) ### Join the Community ![Join the NVIDIA Developer Program](https://developer.download.nvidia.com/icons/m48-developer-1.svg) ### Join the NVIDIA Developer Program * * * ## Ethical AI NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their supporting model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse. For more detailed information on ethical considerations for this model, please see the Model Card++ Explainability, Bias, Safety & Security, and Privacy Subcards. Please report security vulnerabilities or NVIDIA AI Concerns [here](https://www.nvidia.com/en-us/support/submit-security-vulnerability/). * * * ## Porting an app, game, AI tool, or creator workflow to RTX Spark? Tell us about your project. [Submit Form](/local-ai/rtx-spark-app-submit "Submit Form")