Wetron

Parse and visualise neural network models entirely in the browser. ONNX, TFLite, Keras, TorchScript, and ExecuTorch - no server, no upload, no dependencies beyond what the web platform already provides.

Get Started API Reference

Five model formats

ONNX, TFLite, Keras, TorchScript, and ExecuTorch — parsed entirely in the browser with no server required.

Zero server, zero upload

All parsing runs in the browser via native Web APIs. Your model files never leave the device.

React & Svelte renderers

Drop-in graph view and property panel components for React and Svelte, built on @xyflow.