Unable to use gemma 3n onnx model?

[grctest]

System Info

Windows, node.

Environment/Platform

• Website/web-app
• Browser extension
• Server-side (e.g., Node.js, Deno, Bun)
• Desktop app (e.g., Electron)
• Other (e.g., VSCode extension)

Description

bun run .\test.js
20 | constructor(pathOrBuffer, options) {
21 | _OnnxruntimeSessionHandler_inferenceSession.set(this, void 0);
22 | (0, binding_1.initOrt)();
23 | __classPrivateFieldSet(this, _OnnxruntimeSessionHandler_inferenceSession, new binding_1.binding.InferenceSession(), "f");
24 | if (typeof pathOrBuffer === 'string') {
25 | __classPrivateFieldGet(this, _OnnxruntimeSessionHandler_inferenceSession, "f").loadModel(pathOrBuffer, options);
^
error: Deserialize tensor /embed_tokens_per_layer/Gather_quantized failed.tensorprotoutils.cc:1080 onnxruntime::utils::GetExtDataFromTensorProto External initializer: /embed_tokens_per_layer/Gather_quantized offset: 0 size to read: 2013265920 given file_length: 1811281499 are out of bounds or can not be read in full.
at new OnnxruntimeSessionHandler (C:\Users\user\Desktop\git\g3n-electron-webcam-agents\node_modules\onnxruntime-node\dist\backend.js:25:13)
at C:\Users\user\Desktop\git\g3n-electron-webcam-agents\node_modules\onnxruntime-node\dist\backend.js:67:29

Reproduction

Run the example image classification script, using v3.7.0, run into error.