import onnx

from onnxruntime.quantization import quantize_dynamic, QuantType

model_fp32 = 'model.onnx'
model_quant = 'model.quant.onnx'
quantized_model = quantize_dynamic(model_fp32, model_quant)

model_quantized-optimized.onnx

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:b9ed003937f28a4c4f600320cacf2923989f4d2890790a29578b8f79579d6c6d
-size 72506903

model_quantized.onnx

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:451a10638819b01854cac1a95342b783efe93cbf3500faa8fe08a62898d6d75d
-size 72714116
+oid sha256:8b77f140109f46b03f7f4cf5ee8ab0e52441b83a3606f9aab7c64000ff135113
+size 18599309