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 2.37, ONNX Runtime 1.23.0	Download
ONNX	w8a8	Universal	QAIRT 2.37, ONNX Runtime 1.23.0	Download
QNN_DLC	float	Universal	QAIRT 2.42	Download
QNN_DLC	w8a8	Universal	QAIRT 2.42	Download
TFLITE	float	Universal	QAIRT 2.42, TFLite 2.17.0	Download
TFLITE	w8a8	Universal	QAIRT 2.42, TFLite 2.17.0	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® X Elite	1.518 ms	46 - 46 MB	NPU
Inception-v3	ONNX	float	Snapdragon® 8 Gen 3 Mobile	1.26 ms	0 - 122 MB	NPU
Inception-v3	ONNX	float	Qualcomm® QCS8550 (Proxy)	1.66 ms	0 - 54 MB	NPU
Inception-v3	ONNX	float	Qualcomm® QCS9075	2.373 ms	0 - 4 MB	NPU
Inception-v3	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	1.04 ms	0 - 102 MB	NPU
Inception-v3	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	0.917 ms	0 - 103 MB	NPU
Inception-v3	ONNX	w8a8	Snapdragon® X Elite	0.702 ms	24 - 24 MB	NPU
Inception-v3	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.672 ms	0 - 162 MB	NPU
Inception-v3	ONNX	w8a8	Qualcomm® QCS6490	24.544 ms	8 - 41 MB	CPU
Inception-v3	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	0.86 ms	0 - 225 MB	NPU
Inception-v3	ONNX	w8a8	Qualcomm® QCS9075	0.947 ms	0 - 3 MB	NPU
Inception-v3	ONNX	w8a8	Qualcomm® QCM6690	17.95 ms	10 - 20 MB	CPU
Inception-v3	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.557 ms	0 - 123 MB	NPU
Inception-v3	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	13.886 ms	10 - 20 MB	CPU
Inception-v3	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.521 ms	0 - 124 MB	NPU
Inception-v3	QNN_DLC	float	Snapdragon® X Elite	1.426 ms	1 - 1 MB	NPU
Inception-v3	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	1.02 ms	0 - 56 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	7.735 ms	1 - 32 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	1.374 ms	1 - 192 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® SA8775P	2.203 ms	0 - 35 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® QCS9075	2.048 ms	1 - 3 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	2.342 ms	0 - 55 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® SA7255P	7.735 ms	1 - 32 MB	NPU
Inception-v3	QNN_DLC	float	Qualcomm® SA8295P	2.628 ms	0 - 31 MB	NPU
Inception-v3	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	0.849 ms	1 - 35 MB	NPU
Inception-v3	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	0.764 ms	1 - 37 MB	NPU
Inception-v3	QNN_DLC	w8a8	Snapdragon® X Elite	0.693 ms	0 - 0 MB	NPU
Inception-v3	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.48 ms	0 - 77 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® QCS6490	2.29 ms	0 - 2 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	1.527 ms	0 - 47 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	0.629 ms	0 - 2 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® SA8775P	0.779 ms	0 - 51 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® QCS9075	0.62 ms	2 - 4 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® QCM6690	5.798 ms	0 - 50 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	0.896 ms	0 - 74 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® SA7255P	1.527 ms	0 - 47 MB	NPU
Inception-v3	QNN_DLC	w8a8	Qualcomm® SA8295P	1.134 ms	0 - 45 MB	NPU
Inception-v3	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.404 ms	0 - 47 MB	NPU
Inception-v3	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	0.983 ms	0 - 50 MB	NPU
Inception-v3	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.368 ms	0 - 51 MB	NPU
Inception-v3	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1.026 ms	0 - 125 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® QCS8275 (Proxy)	7.782 ms	0 - 73 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® QCS8550 (Proxy)	1.375 ms	0 - 3 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® SA8775P	2.238 ms	0 - 75 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® QCS9075	2.047 ms	0 - 48 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® QCS8450 (Proxy)	2.344 ms	0 - 126 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® SA7255P	7.782 ms	0 - 73 MB	NPU
Inception-v3	TFLITE	float	Qualcomm® SA8295P	2.64 ms	0 - 69 MB	NPU
Inception-v3	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	0.871 ms	0 - 77 MB	NPU
Inception-v3	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	0.772 ms	0 - 77 MB	NPU
Inception-v3	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.484 ms	0 - 73 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® QCS6490	2.265 ms	0 - 25 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	1.616 ms	0 - 47 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	0.644 ms	0 - 2 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® SA8775P	0.852 ms	0 - 50 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® QCS9075	0.699 ms	0 - 25 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® QCM6690	5.787 ms	0 - 49 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	0.945 ms	0 - 75 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® SA7255P	1.616 ms	0 - 47 MB	NPU
Inception-v3	TFLITE	w8a8	Qualcomm® SA8295P	1.251 ms	0 - 45 MB	NPU
Inception-v3	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.432 ms	0 - 50 MB	NPU
Inception-v3	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	1.007 ms	0 - 48 MB	NPU
Inception-v3	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.379 ms	0 - 50 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.