Browser ML
Local browser inference with ONNX and WebAssembly
StyleForge
A browser ML project that runs neural style transfer on the user's device with ONNX Runtime and WebAssembly.
ONNX RuntimeWebAssemblyTypeScriptFrontendImage processing
View source
StyleForge runs neural style transfer in the browser. The image work happens on the user’s device instead of on a backend server.
The app loads an ONNX model, keeps the UI responsive, handles image input and output, and shows progress while WebAssembly runs.
What I built
- ONNX Runtime for portable model inference.
- WebAssembly execution in the browser.
- Image interaction with progress feedback.
- Frontend state for model work that takes time.
Engineering notes
When the user’s device does the work, the app has to be clear about what is happening. StyleForge shows progress, keeps controls responsive, and makes the result easy to compare.