AI PC Development
Web Neural Network API (WebNN)
As machine learning evolves, bridging software and hardware for scalable, web-based solutions has been an ongoing challenge. WebNN is a specification that allows developers to deploy AI entirely within a web browser, and it's being enabled in many popular browsers on Intel® platforms. Web applications gain the ability to create, compile, and run machine learning networks. Use WebNN to take advantage of lower-level acceleration libraries to run AI more efficiently.
Hardware
WebNN can run on the latest Intel® Core™ Ultra processors, but requires a minimum 12th generation Intel Core processor or higher.
How to Install
WebNN requires a compatible browser to run, and Windows* 11 v21H2 (Direct Machine Learning [DirectML] 1.6.0) or higher. Try the latest Google Chrome* Canary or Microsoft Edge* Canary, which requires enabling WebNN functionality in the settings.
|
WebNN Developer Preview
See how you can use WebNN to run ONNX* (Open Neural Network Exchange) models in a browser.
Code Editor
Test out some sample code directly in a compatible browser.
Image Classification
Try image classification with models like MobileNet V2, SqueezeNet, and ResNet V2 50.
Object Detection
Check out this sample for object detection with models like Tiny Yolo V2 and SSD MobileNet V1.
Semantic Segmentation
See how semantic segmentation works with the DeepLab V3 MobileNet V2 model.
Face Recognition
Explore the power of face recognition in your browser with the SSD MobileNet V2 Face model.
Facial Landmark Detection
See how facial landmark detection works with the SSD MobileNet V2 Face model.
Web Neural Network API Explained
Learn about the origins of WebNN, the problems it’s solving, and how it can help you.