The Web Neural Network (WebNN) API: Where we are and what's Next
summary
The WebNN API is a proposed standard for executing neural‑network graphs in browsers as a single integration point across CPU, GPU, and NPU hardware, claiming native‑grade performance without leaving the web. Chromium ships an experimental implementation and ONNX Runtime for the web adds a WebNN execution provider; Windows uses DirectML with vendor drivers (NVIDIA, Intel, AMD, Qualcomm), Apple routes to Core ML, and LiteRT/XNNPACK appears on Android. An in‑browser Stable Diffusion demo runs on Intel discrete GPUs, AMD Radeon RTX 7900 and NVIDIA GeForce RTX 4090 showing measured GPU usage; an origin trial is planned for early next year.
AI-written summary · about
chapters (10)
- 0:00WebNN API introduction
- 1:02Why WebNN API needed
- 3:33What WebNN API is
- 4:36WebNN hardware integration
- 5:20WebNN in web stack
- 6:23DirectML acceleration
- 7:26WebNN status and trials
- 8:08WebNN Stable Diffusion demo
- 9:10Using WebNN API
- 9:52WebNN future plans
related talks
16mThe Web is AI Ready—maximize your AI web development with WebNN
46mThe Ultimate Guide to Browser-Based AI: Privacy-First, Offline-Ready and Future-Proof
45mPrivacy-First In-Browser Generative AI Web Apps
32mPrivacy-first in-browser Generative AI web apps
41mEscaping the Cloud: High-Performance AI in your Browser
30mBeyond the Web of Today
17mRunning Intelligence at the Edge: Web AI in Action
39mThe AI PC and the Era of the Modern Web