Inception-v3: Optimized for Qualcomm Devices

InceptionNetV3 is a machine learning model that can classify images from the Imagenet dataset. It can also be used as a backbone in building more complex models for specific use cases.

This is based on the implementation of Inception-v3 found here. This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.

Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.

Getting Started

There are two ways to deploy this model on your device:

Option 1: Download Pre-Exported Models

Below are pre-exported model assets ready for deployment.

Runtime	Precision	Chipset	SDK Versions	Download
ONNX	float	Universal	QAIRT 2.42, ONNX Runtime 1.25.0	Download
ONNX	w8a8	Universal	QAIRT 2.42, ONNX Runtime 1.25.0	Download
QNN_DLC	float	Universal	QAIRT 2.45	Download
QNN_DLC	w8a8	Universal	QAIRT 2.45	Download
TFLITE	float	Universal	QAIRT 2.45	Download
TFLITE	w8a8	Universal	QAIRT 2.45	Download

For more device-specific assets and performance metrics, visit Inception-v3 on Qualcomm® AI Hub.

Option 2: Export with Custom Configurations

Use the Qualcomm® AI Hub Models Python library to compile and export the model with your own:

• Custom weights (e.g., fine-tuned checkpoints)
• Custom input shapes
• Target device and runtime configurations

This option is ideal if you need to customize the model beyond the default configuration provided here.

See our repository for Inception-v3 on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.image_classification

Model Stats:

• Model checkpoint: Imagenet
• Input resolution: 224x224
• Number of parameters: 23.9M
• Model size (float): 90.9 MB
• Model size (w8a8): 23.3 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
Inception-v3	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	1.071 ms	1 - 43 MB	NPU
Inception-v3	ONNX	float	Snapdragon® X2 Elite	1.134 ms	180 - 180 MB	NPU
Inception-v3	ONNX	float	Snapdragon® X Elite	2.136 ms	180 - 180 MB	NPU
Inception-v3	ONNX	float	Snapdragon® 8 Gen 3 Mobile	1.577 ms	1 - 67 MB	NPU
Inception-v3	ONNX	float	Qualcomm® QCS8550 (Proxy)	2.178 ms	0 - 210 MB	NPU
Inception-v3	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	1.34 ms	0 - 41 MB	NPU
Inception-v3	ONNX	float	Qualcomm® QCS9075	3.709 ms	1 - 46 MB	NPU
Inception-v3	ONNX	float	Qualcomm® QCS8750	1.34 ms	0 - 41 MB	NPU
Inception-v3	ONNX	float	Qualcomm® QCS7181	2.136 ms	180 - 180 MB	NPU
Inception-v3	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.432 ms	0 - 63 MB	NPU
Inception-v3	ONNX	w8a8	Snapdragon® X Elite	0.802 ms	181 - 181 MB	NPU
Inception-v3	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.642 ms	0 - 105 MB	NPU
Inception-v3	ONNX	w8a8	Qualcomm® QCS6490	43.469 ms	18 - 47 MB	CPU
Inception-v3	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	0.851 ms	0 - 173 MB	NPU
Inception-v3	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	26.555 ms	17 - 27 MB	CPU
Inception-v3	ONNX	w8a8	Qualcomm® QCM6690	33.259 ms	16 - 26 MB	CPU
Inception-v3	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.538 ms	0 - 65 MB	NPU
Inception-v3	ONNX	w8a8	Qualcomm® QCS9075	0.9 ms	0 - 45 MB	NPU
Inception-v3	ONNX	w8a8	Qualcomm® QCS7790	26.555 ms	17 - 27 MB	CPU
Inception-v3	ONNX	w8a8	Qualcomm® QCS8750	0.538 ms	0 - 65 MB	NPU
Inception-v3	ONNX	w8a8	Qualcomm® QCS7181	0.802 ms	181 - 181 MB	NPU
Inception-v3	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	0.998 ms	1 - 38 MB	NPU
Inception-v3	QNN_DLC	float	Snapdragon® X2 Elite	1.187 ms	1 - 1 MB	NPU
Inception-v3	QNN_DLC	float	Snapdragon® X Elite	2.069 ms	1 - 1 MB	NPU
Inception-v3	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	1.514 ms	0 - 60 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® QCS8275	13.496 ms	1 - 35 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	1.973 ms	1 - 3 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® SA8775P	3.615 ms	1 - 38 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® SA8650P	3.615 ms	1 - 38 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® SA8255P	3.615 ms	1 - 38 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	3.741 ms	0 - 59 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® SA8295P	4.16 ms	0 - 32 MB	NPU
Inception-v3	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	1.261 ms	1 - 40 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® SA7255P	13.496 ms	1 - 35 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® QCS9075	3.437 ms	1 - 4 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® QCS8750	1.261 ms	1 - 40 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® QCS7181	2.069 ms	1 - 1 MB	NPU
Inception-v3	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.403 ms	0 - 56 MB	NPU
Inception-v3	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.47 ms	0 - 0 MB	NPU
Inception-v3	QNN_DLC	w8a8	Snapdragon® X Elite	0.879 ms	0 - 0 MB	NPU
Inception-v3	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.61 ms	0 - 79 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® QCS6490	3.524 ms	0 - 2 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® QCS8275	2.381 ms	0 - 53 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	0.806 ms	0 - 2 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® SA8775P	1.003 ms	0 - 56 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® SA8650P	1.003 ms	0 - 56 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® SA8255P	1.003 ms	0 - 56 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® SA7255P	2.381 ms	0 - 53 MB	NPU
Inception-v3	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	1.301 ms	0 - 56 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® SA8295P	1.61 ms	0 - 52 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® QCM6690	10.583 ms	0 - 168 MB	NPU
Inception-v3	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.518 ms	0 - 53 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® QCS9075	0.824 ms	0 - 2 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	1.187 ms	0 - 84 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® QCS7790	1.301 ms	0 - 56 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® QCS8750	0.518 ms	0 - 53 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® QCS7181	0.879 ms	0 - 0 MB	NPU
Inception-v3	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	0.993 ms	0 - 68 MB	NPU
Inception-v3	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1.487 ms	0 - 118 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® QCS8275	13.502 ms	0 - 66 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® QCS8550 (Proxy)	2.027 ms	0 - 3 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® SA8775P	3.637 ms	0 - 69 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® SA8650P	3.637 ms	0 - 69 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® SA8255P	3.637 ms	0 - 69 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® QCS8450 (Proxy)	3.745 ms	0 - 120 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® SA8295P	4.145 ms	0 - 64 MB	NPU
Inception-v3	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	1.269 ms	0 - 70 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® SA7255P	13.502 ms	0 - 66 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® QCS9075	3.445 ms	0 - 49 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® QCS8750	1.269 ms	0 - 70 MB	NPU
Inception-v3	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.444 ms	0 - 54 MB	NPU
Inception-v3	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.637 ms	0 - 87 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® QCS6490	3.724 ms	0 - 25 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® QCS8275	2.494 ms	0 - 51 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	0.908 ms	0 - 2 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® SA8775P	1.168 ms	0 - 55 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® SA8650P	1.168 ms	0 - 55 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® SA8255P	1.168 ms	0 - 55 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® SA7255P	2.494 ms	0 - 51 MB	NPU
Inception-v3	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	1.416 ms	0 - 53 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® SA8295P	1.816 ms	0 - 50 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® QCM6690	10.708 ms	0 - 168 MB	NPU
Inception-v3	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.549 ms	0 - 53 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® QCS9075	0.985 ms	0 - 25 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	1.298 ms	0 - 81 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® QCS7790	1.416 ms	0 - 53 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® QCS8750	0.549 ms	0 - 53 MB	NPU

License

• The license for the original implementation of Inception-v3 can be found here.

References

• Rethinking the Inception Architecture for Computer Vision
• Source Model Implementation

Community

• Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
• For questions or feedback please reach out to us.