5 ms·
Seriously I'm looking forward to running ML inference on the web!
by amrb 4y ago
Seriously I'm looking forward to running ML inference on the web!
- fulafel 4y agoPeople have been doing it for long with WebGL, see eg https://github.com/tensorflow/tfjs https://github.com/tensorflow/tfjs and https://cloudblogs.microsoft.com/opensource/2021/09/02/onnx-runtime-web-running-your-machine-learning-model-in-browser/ https://cloudblogs.microsoft.com/opensource/2021/09/02/onnx-...
- paulgb 4y agoIt will be interesting to see the performance differential. Tensorflow.js provides a benchmark tool[1]. When I ran them[2] on an M1 MacBook Pro, WebGPU (in Chrome Canary) was usually 2x as fast as WebGL on large models, sometimes 3x. [1] https://tfjs-benchmarks.web.app/local-benchmark/ https://tfjs-benchmarks.web.app/local-benchmark/ [2] https://digest.browsertech.com/archive/browsertech-digest-what-webgpu-means-for-games/ https://digest.browsertech.com/archive/browsertech-digest-wh...