Realize your vision with a guide to building Windows on Snapdragon
summary
Qualitative claims become concrete here: Qualcomm ships a Snapdragon Dev Kit with 12‑core CPUs, Hexagon NPU, and 32 GB RAM that can hit 40 TOPS for on‑device models; McAfee ports a real‑time deep‑fake audio detector to the NPU, running inference locally to preserve privacy, cut latency, and lower power. Useful tooling includes Snapdragon LLVM, Qualcomm Profiler, AI Hub, ONNX→DLC→QAIT conversion and AI Engine Direct for model optimization; documentation and samples live at developer.qualcomm.com.
AI-written summary · about
chapters (9)
- 0:00Windows on Snapdragon intro
- 1:12Why port apps to Snapdragon
- 2:38McAfee’s native porting
- 4:23Porting process and NPU
- 5:26Qualcomm’s tools and resources
- 7:38Local AI processing benefits
- 9:45New AI developer features
- 12:08AI integration advice
- 13:56Snapdragon Dev Kit
related talks
39mQualcomm Technologies and the power of Windows AI Q&A
54mHarnessing On-Device AI to Transform the Smartphone Experience
17mOverview of Qualcomm Snapdragon Neural Processing Engine
29mQualcomm® AI Stack for developers and extension to on-device AI
14mAccelerating On-Device AI Development
32mDeploy AI with Simulink
08mUnlocking AI at Scale for PyTorch Developers
26mHow to use Qualcomm AI Hub