The Chrome team recently released WebGPU, a major upgrade for the web that enables high-performance 3D graphics and data-parallel computation. This API was developed by the W3C's "GPU for the Web" Community Group, and its cross-platform compatibility is set to revolutionize the graphic API space. With the potential to unlock new app development opportunities, WebGPU is particularly exciting for AI and web graphics. Enabling WebGPU on a Macbook Pro M1 with macOS Ventura is easy: simply download Chrome Beta, and enable WebGPU using chrome://flags/#enable-unsafe-webgpu.
Check out the demo video, blog posts, and documentation to explore the possibilities of WebGPU, including the ability to run AI models with TensorFlow.js.
Check out the demo video, blog posts, and documentation to explore the possibilities of WebGPU, including the ability to run AI models with TensorFlow.js.
WebGPU playgrounds https://toji.github.io/webgpu-clustered-shading/ https://webgpu.github.io/webgpu-samples/ Documentation Get started with GPU Compute on the web https://developer.chrome.com/articles/gpu-compute/ WebGPU W3C https://gpuweb.github.io/gpuweb/ Technical introduction to the WebGPU Shading Language https://google.github.io/tour-of-wgsl/ WebGPU also allows running AI models. For instance, with TensorFlow.js you can test AI models like BlazeFace, BodyPix, PoseDetection, AutoML Image classification and more with the package below. GPU accelerated WebGPU backend for TensorFlow.js https://www.npmjs.com/package/@tensorflow/tfjs-backend-webgp...
#WebGPU #GraphicsProgramming #AI