--- markdown_generated_at: "2026-11-10" title: "Build and ship On-Device AI" url: "https://www.qualcomm.com/developer/artificial-intelligence" --- # Build and ship On-Device AI ![](https://s7d1.scene7.com/is/image/dmqualcommprod/header-1440-x2?fmt=webp-alpha&qlt=85) Deploy agentic and generative AI models directly with Snapdragon® and Qualcomm Dragonwing™ processors. Optimize for inference latency, cost, and power efficiency. [Get started](https://docs.qualcomm.com/doc/80-70018-15B/topic/introduction.html) [View AI capabilities](https://www.qualcomm.com/artificial-intelligence) ## Build with your preferred framework, deploy with Qualcomm Develop models in TensorFlow, PyTorch, ONNX, or LiteRT and run them with hardware-accelerated inference across over four billion devices powered by Snapdragon and Dragonwing processors for mobile, compute, and IoT. ![](https://s7d1.scene7.com/is/image/dmqualcommprod/framework-desktop-x2?fmt=webp-alpha&qlt=85) ![](https://s7d1.scene7.com/is/image/dmqualcommprod/1440-x2-black?fmt=webp-alpha&qlt=85) ## Choose your developer approach Start with the Guided approach for pre-built workflows and framework runtimes or choose the Native approach for direct control over model execution, optimization, and runtime targeting. ![](https://s7d1.scene7.com/is/image/dmqualcommprod/frame-purple?fmt=webp-alpha&qlt=85&wid=1800&fit=fit,1) ### Guided Get started quickly with guided workflows and runtimes for ML frameworks. ![](https://s7d1.scene7.com/is/image/dmqualcommprod/frame-blue?fmt=webp-alpha&qlt=85&wid=1800&fit=fit,1) ### Native Native tools provide more direct control over optimization, runtime behavior, and model execution. ## AI developer workflow #### 1. Acquire Training Data [Sample Datasets](https://www.qualcomm.com/developer/artificial-intelligence/datasets) Sample datasets for training models across research, prototyping, and experimentation. [Datasets - Hugging Face](https://huggingface.co/datasets) Access datasets for your use case on Hugging Face. #### 2. Start with models Download a pre-optimized model or bring your own and prepare it for on-device execution. ![](https://s7d1.scene7.com/is/image/dmqualcommprod/frame-purple?fmt=webp-alpha&qlt=85&wid=1800&fit=fit,1) GUIDED [Pre-optimized models from Qualcomm AI Hub](https://aihub.qualcomm.com/models) Hundreds of pre-optimized ML models, Small Language Models (SLMs), and Large Language Models (LLMs) are ready to profile, test, and deploy. [Models on Hugging Face](https://huggingface.co/models) Access models for your use case on Hugging Face. ![](https://s7d1.scene7.com/is/image/dmqualcommprod/frame-blue?fmt=webp-alpha&qlt=85&wid=1800&fit=fit,1) NATIVE [Qualcomm AI Runtime SDK (QAIRT)](https://www.qualcomm.com/developer/software/qualcomm-ai-runtime-sdk-qairt) Prepare models from your own training data using Qualcomm® AI Runtime SDK (QAIRT). QAIRT includes conversion and optimization tools for TensorFlow, PyTorch, and ONNX, with execution across CPU, GPU, NPU, and Qualcomm® Sensing Hub on Snapdragon and Dragonwing. Related Resource [Building and Executing Your Model tutorial](https://docs.qualcomm.com/nav/home/building_and_executing_tutorial.html?product=1601111740010412) which includes steps to build a model from data. [Qualcomm Neural Processing SDK](https://www.qualcomm.com/developer/software/neural-processing-sdk-for-ai) If you want to prepare a model from your own training data, start with the Qualcomm Neural Processing SDK which includes tools to build a compatible model for execution on Snapdragon or Dragonwing. #### 3. Choose Device Choose from an array of devices with Snapdragon and Dragonwing for your compute, IoT, or mobile projects. ![](https://s7d1.scene7.com/is/image/dmqualcommprod/frame-purple?fmt=webp-alpha&qlt=85&wid=1800&fit=fit,1) GUIDED [Qualcomm® Device Cloud](https://qdc.qualcomm.com/) Access real compute and mobile devices with Dragonwing and Snapdragon remotely through your browser. Build, test, debug, and profile on physical devices without requiring physical devices. [Qualcomm AI Hub](https://aihub.qualcomm.com/get-started#workbench) Use Qualcomm AI Hub to profile and test on devices with Qualcomm AI Hub Workbench. ![](https://s7d1.scene7.com/is/image/dmqualcommprod/frame-blue?fmt=webp-alpha&qlt=85&wid=1800&fit=fit,1) NATIVE [Windows on Snapdragon / Compute](https://www.qualcomm.com/snapdragon/laptops-and-tablets) Snapdragon X Series processors power the fastest, most power-efficient AI PCs from several major brands. [Mobile](https://www.qualcomm.com/snapdragon/smartphones) Explore the full range of Android OS devices powered by Snapdragon processors for your mobile apps and games. [IoT and Robotics](https://www.qualcomm.com/developer/iot) Choose from powerful evaluation kit (EVK) options for your next agentic AI or GenAI project. #### 4. Optimize for on device Convert your models into Snapdragon and Dragonwing compatible formats and optimize them for size, accuracy, and performance during execution. ![](https://s7d1.scene7.com/is/image/dmqualcommprod/frame-purple?fmt=webp-alpha&qlt=85&wid=1800&fit=fit,1) GUIDED [Qualcomm AI Hub](https://aihub.qualcomm.com/get-started#workbench) Use Qualcomm AI Hub Workbench to quickly optimize a trained model for execution on over 60 cloud-based devices using supported runtimes. [Edge Impulse](https://www.qualcomm.com/developer/software/edge-impulse) Use Edge Impulse to build IoT datasets, train models, and optimize libraries to run directly on devices with Dragonwing. ![](https://s7d1.scene7.com/is/image/dmqualcommprod/frame-blue?fmt=webp-alpha&qlt=85&wid=1800&fit=fit,1) NATIVE [AI Model Efficiency Toolkit (AIMET)](https://github.com/qualcomm/aimet) Use AIMET to quantize and compress models trained with PyTorch or ONNX before running them with ONNX Runtime (ORT), Qualcomm Neural Processing SDK, or Qualcomm AI Engine Direct SDK (QNN). [Qualcomm AI Runtime SDK (QAIRT)](https://www.qualcomm.com/developer/software/qualcomm-ai-runtime-sdk-qairt) Qualcomm AI Runtime SDK (QAIRT) includes several analysis and optimization tools to prepare a model for execution using the SDK's runtime. Related Resource [Quantization code samples introductory tutorial](https://docs.qualcomm.com/nav/home/building_and_executing_tutorial_linux_host.html?product=1601111740010412) #### 5. Configure a Runtime Below are several runtimes you can use for hardware-accelerated inference in your application running on Snapdragon or Dragonwing. ![](https://s7d1.scene7.com/is/image/dmqualcommprod/frame-purple?fmt=webp-alpha&qlt=85&wid=1800&fit=fit,1) GUIDED [Qualcomm AI Hub](https://aihub.qualcomm.com/get-started) Choose one of the many sample applications as your starting point for loading a model and running inference. See the Get Started with Qualcomm AI Hub Apps tutorial [AI/ML Framework Runtimes](https://www.qualcomm.com/developer/software/qualcomm-ai-engine-direct-sdk) Built on the Qualcomm AI Engine Direct SDK, these runtime integrations let you use your preferred AI/ML and Generative AI framework. [ONNX Runtime (ORT)](https://github.com/onnxruntime/onnxruntime-qnn) For ONNX models, ONNX Runtime (ORT) includes a Qualcomm Plugin Execution Provider that you can incorporate into your application. [LiteRT](https://docs.qualcomm.com/doc/80-63442-10/topic/index_TfLite.html) To run LiteRT models incorporate the LiteRT Delegate into your application. [ExecuTorch](https://docs.pytorch.org/executorch/1.2/backends-qualcomm.html) For PyTorch models, incorporate ExecuTorch into your application that includes a Qualcomm® Hexagon™ Delegate. ![](https://s7d1.scene7.com/is/image/dmqualcommprod/frame-blue?fmt=webp-alpha&qlt=85&wid=1800&fit=fit,1) NATIVE [Qualcomm AI Runtime SDK (QAIRT)](https://www.qualcomm.com/developer/software/qualcomm-ai-runtime-sdk-qairt) Qualcomm AI Runtime SDK (QAIRT) is a suite of tools and runtimes to build, optimize, and run ML models generated by TensorFlow, PyTorch, and ONNX. Related Resource [Building and Executing Your Model tutorial](https://docs.qualcomm.com/nav/home/building_and_executing_tutorial.html?product=1601111740010412) which includes steps to build a model from data. [Sample App Tutorial](https://docs.qualcomm.com/nav/home/tutorial_sample_app.html?product=924033590759186372) #### 6. Deploy and run Deploy your app, your model, or both and choose the tools that match your pipeline. ![](https://s7d1.scene7.com/is/image/dmqualcommprod/frame-purple?fmt=webp-alpha&qlt=85&wid=1800&fit=fit,1) GUIDED [Qualcomm AI Hub](https://aihub.qualcomm.com/get-started) Choose one of the many sample applications as your starting point for loading a model and running inference. See the Get Started with Qualcomm AI Hub Apps tutorial [FoundriesFactory™ Platform](https://www.qualcomm.com/developer/software/foundriesfactory) Use FoundriesFactory to build a CI/CD pipeline that manages OTA updates for apps and models to IoT edge devices in the field via containerization. ![](https://s7d1.scene7.com/is/image/dmqualcommprod/frame-blue?fmt=webp-alpha&qlt=85&wid=1800&fit=fit,1) NATIVE [Qualcomm AI Runtime SDK (QAIRT)](https://www.qualcomm.com/developer/software/qualcomm-ai-runtime-sdk-qairt) Qualcomm AI Runtime SDK (QAIRT) is a suite of tools and runtimes to build, optimize, and run ML models generated by TensorFlow, PyTorch, and ONNX. ## Use Cases ![](https://s7d1.scene7.com/is/image/dmqualcommprod/use-case-iot?fmt=webp-alpha&qlt=85&wid=1800&fit=fit,1) ### IoT - Vision language models to assess remote areas - Object analysis for healthcare or retail - Predictive maintenance in factories - Precision agriculture to monitor crops and soil conditions [Learn more](https://www.qualcomm.com/developer/iot#deploywithai) ![](https://s7d1.scene7.com/is/image/dmqualcommprod/use-case-robotics?fmt=webp-alpha&qlt=85&wid=1800&fit=fit,1) ### Robotics - Factory robots on assembly lines - Voice-controlled robotic arms - Autonomous restaurant servers - Autonomous drones - Automated healthcare assistants [Learn more](https://www.qualcomm.com/internet-of-things/applications/robotics-processors) ![](https://s7d1.scene7.com/is/image/dmqualcommprod/use-case-compute?fmt=webp-alpha&qlt=85&wid=1800&fit=fit,1) ### Compute - Fast photo and video editing - Turn agentic IT ticket overload into a daily insight pipeline - Voice-controlled user interfaces - Multi-device agentic orchestration [Learn more](https://www.qualcomm.com/developer/windows-on-snapdragon/windows-on-snapdragon-ai) ![](https://s7d1.scene7.com/is/image/dmqualcommprod/use-cases-mobile?fmt=webp-alpha&qlt=85&wid=1800&fit=fit,1) ### Mobile - Chatbots for documents and images - Turn medical records into simple conversations - Agentic web search, deep research and automation workflows - On-demand translation and image generation [Learn more](https://www.qualcomm.com/) ![](https://s7d1.scene7.com/is/image/dmqualcommprod/use-case-gaming?fmt=webp-alpha&qlt=85&wid=1800&fit=fit,1) ### Gaming - Intelligent and conversational non-playable characters (NPCs) - Agent-based teammates - In-game assistants - Dynamic graphics or stories [Learn more](https://www.qualcomm.com/developer/snapdragon-game-toolkit) ![](https://s7d1.scene7.com/is/image/dmqualcommprod/use-case-vr?fmt=webp-alpha&qlt=85&wid=1800&fit=fit,1) ### XR - Context-aware personal assistants - Spatial perception and scene understanding - Digital twins for enterprise and industrial environments - Immersive training and education for teams [Learn more](https://www.qualcomm.com/developer/software/snapdragon-spaces-compatibility-plugin-for-android-xr) ## Discover our latest insights OnQ Blog ### [AI on Snapdragon is AI that moves with you](https://www.qualcomm.com/news/onq/2026/11/ai-on-snapdragon-agentic-age) Nov 9, 2026 | 6 min read [Snapdragon](https://www.qualcomm.com/news/onq?tags=Snapdragon) [AI](https://www.qualcomm.com/news/onq?tags=AI) [Agentic AI](https://www.qualcomm.com/news/onq?tags=Agentic%20AI) OnQ Blog ### [Qualcomm HBC demonstrates AI inference leadership with validated silicon](https://www.qualcomm.com/news/onq/2026/11/qualcomm-hbc-validated-silicon-ai-inference) Nov 2, 2026 | 3 min read [Dragonfly](https://www.qualcomm.com/news/onq?tags=Dragonfly) [HBC](https://www.qualcomm.com/news/onq?tags=HBC) [AI Inference](https://www.qualcomm.com/news/onq?tags=AI%20Inference) Developer Blog ### [How to write fast AArch64 SVE/SME Kernels with Ripple and LLVM](https://www.qualcomm.com/developer/blog/2026/10/ripple-llvm-aarch64-sve-sme-kernel-implementation) Oct 29, 2026 [LLVM](https://www.qualcomm.com/developer/blog?tags=LLVM) [Open Source](https://www.qualcomm.com/developer/blog?tags=Open%20Source) Developer Blog ### [Same-day upstream Linux support now posted for Snapdragon 8 Elite Gen 6 and Snapdragon Elite Extreme Gen 6 mobile platforms](https://www.qualcomm.com/developer/blog/2026/10/same-day-snapdragon-8-elite-gen-6-upstream-linux-support) Oct 27, 2026 [Adreno](https://www.qualcomm.com/developer/blog?tags=Adreno) [Linux](https://www.qualcomm.com/developer/blog?tags=Linux) [Snapdragon](https://www.qualcomm.com/developer/blog?tags=Snapdragon) Developer Blog ### [How to share Vulkan and OpenCL workloads on Android with Qualcomm Adreno GPUs](https://www.qualcomm.com/developer/blog/2026/10/how-to-share-vulkan-opencl-workloads-android-adreno-gpus) Oct 22, 2026 [Vulkan](https://www.qualcomm.com/developer/blog?tags=Vulkan) [OpenCL](https://www.qualcomm.com/developer/blog?tags=OpenCL) [Android](https://www.qualcomm.com/developer/blog?tags=Android) Developer Blog ### [How to Install OpenClaw and Hermes Agent on Qualcomm Arduino Boards, Rubik Pi 3 and PCs with Snapdragon](https://www.qualcomm.com/developer/blog/2026/04/install-openclaw-hermes-agent-qualcomm-arduino-rubik-pi-snapdragon-pcs) Apr 24, 2026 [AI](https://www.qualcomm.com/developer/blog?tags=AI) [Linux](https://www.qualcomm.com/developer/blog?tags=Linux) [Windows on Snapdragon](https://www.qualcomm.com/developer/blog?tags=Windows%20on%20Snapdragon) Developer Blog ### [Comparing LLM models with the Qualcomm AI Inference Suite](https://www.qualcomm.com/developer/blog/2026/02/comparing-llm-models-with-qualcomm-ai-inference-suite) Feb 11, 2026 [AI](https://www.qualcomm.com/developer/blog?tags=AI) [Cloud](https://www.qualcomm.com/developer/blog?tags=Cloud) [Generative AI](https://www.qualcomm.com/developer/blog?tags=Generative%20AI) ## Connect with our communities [LinkedIn](https://www.linkedin.com/showcase/qualcomm-developer) [See what's new](https://x.com/Qualcomm_Dev) [YouTube](https://www.youtube.com/qualcommdev) [GitHub](https://github.com/qualcomm) ![](https://s7d1.scene7.com/is/image/dmqualcommprod/discord-icon-1?$QC_Responsive$&fmt=png-alpha) [Discord](https://discord.com/invite/qualcommdevelopernetwork) [Visit the forum](https://mysupport.qualcomm.com/supportforums/s/) ![](https://s7d1.scene7.com/is/image/dmqualcommprod/ai-hub-bg-desktop?fmt=webp-alpha&qlt=85)