PyTorch's API that trains in a browser tab on WebGPU — autograd, nn, optim, capture/replay, int8, ONNX out; held to real torch's values on 4,764 golden cases; measured on one page against TF.js, jax-js, Burn and ONNX Runtime Web (4–12× faster training, ahead of ORT at inference)
training benchmark machine-learning typescript browser deep-learning neural-network pytorch autograd tensor webgpu onnx pyodide wgsl jupyterlite in-browser-ml webgpu-compute in-browser-training
-
Updated
Oct 3, 2026 - Python