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 2.50, ONNX Runtime 1.30.0	Download
ONNX	w8a8	Universal	QAIRT 2.50, ONNX Runtime 1.30.0	Download
QNN_DLC	float	Universal	QAIRT 2.50	Download
QNN_DLC	w8a8	Universal	QAIRT 2.50	Download
TFLITE	float	Universal	QAIRT 2.50	Download
TFLITE	w8a8	Universal	QAIRT 2.50	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:

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

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 For Galaxy Mobile	1.081 ms	0 - 45 MB	NPU
Inception-v3	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	1.295 ms	1 - 46 MB	NPU
Inception-v3	ONNX	float	Snapdragon® X2 Elite	1.126 ms	1 - 1 MB	NPU
Inception-v3	ONNX	float	Snapdragon® X Elite	2.168 ms	46 - 46 MB	NPU
Inception-v3	ONNX	float	Snapdragon® 8 Gen 3 Mobile	1.552 ms	0 - 63 MB	NPU
Inception-v3	ONNX	float	Snapdragon® 8 Gen 1 Mobile	3.665 ms	0 - 66 MB	NPU
Inception-v3	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	4.187 ms	1 - 6 MB	NPU
Inception-v3	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	2.193 ms	0 - 49 MB	NPU
Inception-v3	ONNX	float	Qualcomm® QCS8450	3.665 ms	0 - 66 MB	NPU
Inception-v3	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	3.677 ms	1 - 5 MB	NPU
Inception-v3	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	2.168 ms	46 - 46 MB	NPU
Inception-v3	ONNX	float	Qualcomm® Dragonwing™ Q-8750	1.295 ms	1 - 46 MB	NPU
Inception-v3	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	0.411 ms	0 - 70 MB	NPU
Inception-v3	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.526 ms	0 - 64 MB	NPU
Inception-v3	ONNX	w8a8	Snapdragon® X2 Elite	0.357 ms	1 - 1 MB	NPU
Inception-v3	ONNX	w8a8	Snapdragon® X Elite	0.789 ms	24 - 24 MB	NPU
Inception-v3	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.61 ms	0 - 109 MB	NPU
Inception-v3	ONNX	w8a8	Snapdragon® 8 Gen 1 Mobile	0.992 ms	0 - 108 MB	NPU
Inception-v3	ONNX	w8a8	Qualcomm® Dragonwing™ QCS6490	3.107 ms	0 - 4 MB	NPU
Inception-v3	ONNX	w8a8	Qualcomm® Dragonwing™ IQ-8275	0.963 ms	0 - 4 MB	NPU
Inception-v3	ONNX	w8a8	Qualcomm® Dragonwing™ QCS8550 (Proxy)	0.832 ms	0 - 2 MB	NPU
Inception-v3	ONNX	w8a8	Qualcomm® QCS8450	0.992 ms	0 - 108 MB	NPU
Inception-v3	ONNX	w8a8	Qualcomm® Dragonwing™ IQ-9075	0.919 ms	0 - 3 MB	NPU
Inception-v3	ONNX	w8a8	Qualcomm® Dragonwing™ IQ-X7181	0.789 ms	24 - 24 MB	NPU
Inception-v3	ONNX	w8a8	Qualcomm® Dragonwing™ Q-6690	10.679 ms	0 - 179 MB	NPU
Inception-v3	ONNX	w8a8	Qualcomm® Dragonwing™ Q-7790	1.333 ms	0 - 68 MB	NPU
Inception-v3	ONNX	w8a8	Qualcomm® Dragonwing™ Q-8750	0.526 ms	0 - 64 MB	NPU
Inception-v3	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	1.333 ms	0 - 68 MB	NPU
Inception-v3	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	1.004 ms	1 - 40 MB	NPU
Inception-v3	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	1.233 ms	0 - 36 MB	NPU
Inception-v3	QNN_DLC	float	Snapdragon® X2 Elite	1.171 ms	1 - 1 MB	NPU
Inception-v3	QNN_DLC	float	Snapdragon® X Elite	2.054 ms	1 - 1 MB	NPU
Inception-v3	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	1.454 ms	1 - 64 MB	NPU
Inception-v3	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	3.414 ms	0 - 63 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	4.002 ms	1 - 5 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	1.953 ms	1 - 3 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® SA8775P	3.581 ms	1 - 40 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® SA8650P	3.581 ms	1 - 40 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® SA8255P	3.581 ms	1 - 40 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® QCS8450	3.414 ms	0 - 63 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	3.428 ms	1 - 4 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	2.054 ms	1 - 1 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	1.233 ms	0 - 36 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® SA7255P	13.494 ms	1 - 38 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® SA8295P	4.055 ms	0 - 36 MB	NPU
Inception-v3	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	0.416 ms	0 - 58 MB	NPU
Inception-v3	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.512 ms	0 - 57 MB	NPU
Inception-v3	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.458 ms	0 - 0 MB	NPU
Inception-v3	QNN_DLC	w8a8	Snapdragon® X Elite	0.876 ms	0 - 0 MB	NPU
Inception-v3	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.581 ms	0 - 83 MB	NPU
Inception-v3	QNN_DLC	w8a8	Snapdragon® 8 Gen 1 Mobile	0.962 ms	0 - 84 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® Dragonwing™ QCS6490	3.133 ms	0 - 3 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® Dragonwing™ IQ-8275	0.928 ms	0 - 3 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® Dragonwing™ QCS8550 (Proxy)	0.801 ms	0 - 2 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® SA8775P	0.966 ms	0 - 62 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® SA8650P	0.966 ms	0 - 62 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® SA8255P	0.966 ms	0 - 62 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® QCS8450	0.962 ms	0 - 84 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® Dragonwing™ IQ-9075	0.874 ms	0 - 3 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® Dragonwing™ IQ-X7181	0.876 ms	0 - 0 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® Dragonwing™ Q-6690	10.519 ms	0 - 166 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® Dragonwing™ Q-7790	1.299 ms	0 - 60 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® Dragonwing™ Q-8750	0.512 ms	0 - 57 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® SA7255P	2.378 ms	0 - 57 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® SA8295P	1.468 ms	0 - 55 MB	NPU
Inception-v3	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	1.299 ms	0 - 60 MB	NPU
Inception-v3	TFLITE	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	1.005 ms	0 - 72 MB	NPU
Inception-v3	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	1.239 ms	0 - 68 MB	NPU
Inception-v3	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1.503 ms	0 - 117 MB	NPU
Inception-v3	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	3.404 ms	0 - 124 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	4.028 ms	0 - 50 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	2.025 ms	0 - 3 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® SA8775P	3.594 ms	0 - 71 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® SA8650P	3.594 ms	0 - 71 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® SA8255P	3.594 ms	0 - 71 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® QCS8450	3.404 ms	0 - 124 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	3.421 ms	0 - 49 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	1.239 ms	0 - 68 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® SA8295P	4.053 ms	0 - 67 MB	NPU
Inception-v3	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	0.448 ms	0 - 55 MB	NPU
Inception-v3	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.543 ms	0 - 60 MB	NPU
Inception-v3	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.627 ms	0 - 84 MB	NPU
Inception-v3	TFLITE	w8a8	Snapdragon® 8 Gen 1 Mobile	1.08 ms	0 - 92 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® Dragonwing™ QCS6490	3.748 ms	0 - 26 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® Dragonwing™ IQ-8275	0.974 ms	0 - 26 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® Dragonwing™ QCS8550 (Proxy)	0.886 ms	0 - 2 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® SA8775P	1.155 ms	0 - 60 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® SA8650P	1.155 ms	0 - 60 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® SA8255P	1.155 ms	0 - 60 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® QCS8450	1.08 ms	0 - 92 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® Dragonwing™ IQ-9075	1.032 ms	0 - 26 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® Dragonwing™ Q-6690	10.571 ms	0 - 166 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® Dragonwing™ Q-7790	1.415 ms	0 - 58 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® Dragonwing™ Q-8750	0.543 ms	0 - 60 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® SA8295P	1.631 ms	0 - 54 MB	NPU
Inception-v3	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	1.415 ms	0 - 58 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.