Wasm, WebGPU, & WebNN: How compute abstraction are enabling client side AI
summary
Wasm compiles C++ Python and other languages to run on the CPU with threads via web workers and SIMD, enabling things like llama.cpp and compiled Python SDKs; WebGPU adds compute shaders for high GPU throughput and is landing broadly in browsers (except Firefox mobile around release time); WebNN is early but promises OS-backed CPU GPU NPU routing via Windows ML Core ML etc; expect fast client media processing, Whisper on WebGPU, OCR and small-model text classification, but watch bundle size and update responsibility.
AI-written summary · about
chapters (8)
- 0:00Compute abstractions intro
- 1:50WebAssembly for CPU
- 3:13WebGPU compute shaders
- 4:20WebNN access
- 6:25Why compute abstractions for AI
- 8:08Advantages over built-in APIs
- 9:52AI use cases
- 13:01Q&A
related talks
30mBeyond the Web of Today
16mThe Web is AI Ready—maximize your AI web development with WebNN
39mThe AI PC and the Era of the Modern Web
25mLeveraging Wasm for Portable AI Inference Across GPUs, CPUs, OS & Cloud-Nativ...
41mEscaping the Cloud: High-Performance AI in your Browser
38mWebGPU: Next-generation 3D Graphics on the Web
01h 42mWebAssembly: Writing Once and Running Everywhere NG
31mRunning AI workloads with WASM - is it production ready? [Panel]