ResNeXt101: Optimized for Qualcomm Devices

ResNeXt101 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 ResNeXt101 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 1.24.3	Download
ONNX	w8a8	Universal	QAIRT 2.42, ONNX Runtime 1.24.3	Download
QNN_DLC	float	Universal	QAIRT 2.43	Download
QNN_DLC	w8a8	Universal	QAIRT 2.43	Download
TFLITE	float	Universal	QAIRT 2.43, TFLite 2.19.1	Download
TFLITE	w8a8	Universal	QAIRT 2.43, TFLite 2.19.1	Download

For more device-specific assets and performance metrics, visit ResNeXt101 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 ResNeXt101 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: 88.7M
• Model size (float): 338 MB
• Model size (w8a8): 87.3 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
ResNeXt101	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	3.043 ms	1 - 203 MB	NPU
ResNeXt101	ONNX	float	Snapdragon® X2 Elite	3.089 ms	173 - 173 MB	NPU
ResNeXt101	ONNX	float	Snapdragon® X Elite	6.747 ms	172 - 172 MB	NPU
ResNeXt101	ONNX	float	Snapdragon® 8 Gen 3 Mobile	4.528 ms	0 - 387 MB	NPU
ResNeXt101	ONNX	float	Qualcomm® QCS8550 (Proxy)	6.603 ms	0 - 196 MB	NPU
ResNeXt101	ONNX	float	Qualcomm® QCS9075	9.734 ms	0 - 4 MB	NPU
ResNeXt101	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	3.874 ms	0 - 173 MB	NPU
ResNeXt101	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	1.564 ms	0 - 216 MB	NPU
ResNeXt101	ONNX	w8a8	Snapdragon® X2 Elite	1.386 ms	87 - 87 MB	NPU
ResNeXt101	ONNX	w8a8	Snapdragon® X Elite	3.14 ms	87 - 87 MB	NPU
ResNeXt101	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	2.119 ms	0 - 255 MB	NPU
ResNeXt101	ONNX	w8a8	Qualcomm® QCS6490	112.53 ms	7 - 43 MB	CPU
ResNeXt101	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	2.908 ms	0 - 100 MB	NPU
ResNeXt101	ONNX	w8a8	Qualcomm® QCS9075	3.143 ms	0 - 3 MB	NPU
ResNeXt101	ONNX	w8a8	Qualcomm® QCM6690	73.678 ms	0 - 12 MB	CPU
ResNeXt101	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.92 ms	0 - 224 MB	NPU
ResNeXt101	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	68.798 ms	0 - 12 MB	CPU
ResNeXt101	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	3.087 ms	1 - 193 MB	NPU
ResNeXt101	QNN_DLC	float	Snapdragon® X2 Elite	3.737 ms	1 - 1 MB	NPU
ResNeXt101	QNN_DLC	float	Snapdragon® X Elite	6.925 ms	1 - 1 MB	NPU
ResNeXt101	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	4.68 ms	0 - 369 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	35.92 ms	1 - 192 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	6.745 ms	1 - 4 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® SA8775P	10.307 ms	1 - 203 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® QCS9075	9.972 ms	1 - 3 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	10.019 ms	0 - 313 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® SA7255P	35.92 ms	1 - 192 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® SA8295P	10.922 ms	1 - 140 MB	NPU
ResNeXt101	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	4.029 ms	0 - 189 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	1.513 ms	0 - 209 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® X2 Elite	1.766 ms	0 - 0 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® X Elite	3.057 ms	0 - 0 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	2.164 ms	0 - 251 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® QCS6490	9.175 ms	0 - 2 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	6.508 ms	0 - 206 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	2.838 ms	0 - 35 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® SA8775P	3.489 ms	0 - 208 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® QCS9075	3.133 ms	0 - 2 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® QCM6690	31.837 ms	0 - 370 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	3.904 ms	0 - 250 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® SA7255P	6.508 ms	0 - 206 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® SA8295P	4.25 ms	0 - 209 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.846 ms	0 - 205 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	4.07 ms	0 - 238 MB	NPU
ResNeXt101	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	3.093 ms	0 - 400 MB	NPU
ResNeXt101	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	4.62 ms	0 - 572 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® QCS8275 (Proxy)	35.89 ms	0 - 396 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® QCS8550 (Proxy)	6.808 ms	0 - 3 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® SA8775P	10.303 ms	0 - 396 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® QCS9075	10.038 ms	0 - 174 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® QCS8450 (Proxy)	9.908 ms	0 - 511 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® SA7255P	35.89 ms	0 - 396 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® SA8295P	10.999 ms	0 - 336 MB	NPU
ResNeXt101	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	3.946 ms	0 - 380 MB	NPU
ResNeXt101	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	1.477 ms	0 - 213 MB	NPU
ResNeXt101	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	2.003 ms	0 - 252 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® QCS6490	9.187 ms	0 - 88 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	6.321 ms	0 - 212 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	2.669 ms	0 - 357 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® SA8775P	3.354 ms	0 - 213 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® QCS9075	2.954 ms	0 - 89 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® QCM6690	29.872 ms	0 - 372 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	3.746 ms	0 - 252 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® SA7255P	6.321 ms	0 - 212 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® SA8295P	4.11 ms	0 - 214 MB	NPU
ResNeXt101	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.774 ms	0 - 208 MB	NPU
ResNeXt101	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	3.97 ms	0 - 229 MB	NPU

License

• The license for the original implementation of ResNeXt101 can be found here.

References

• Aggregated Residual Transformations for Deep Neural Networks
• 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.