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 2.45	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 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	3.048 ms	1 - 203 MB	NPU
ResNeXt101	ONNX	float	Snapdragon® 8 Elite Mobile	3.937 ms	0 - 212 MB	NPU
ResNeXt101	ONNX	float	Snapdragon® X2 Elite	3.121 ms	173 - 173 MB	NPU
ResNeXt101	ONNX	float	Snapdragon® X Elite	6.716 ms	172 - 172 MB	NPU
ResNeXt101	ONNX	float	Snapdragon® X Elite	6.716 ms	172 - 172 MB	NPU
ResNeXt101	ONNX	float	Snapdragon® 8 Gen 3 Mobile	4.496 ms	0 - 372 MB	NPU
ResNeXt101	ONNX	float	Qualcomm® QCS8550 (Proxy)	6.603 ms	0 - 195 MB	NPU
ResNeXt101	ONNX	float	Qualcomm® QCS9075	9.733 ms	0 - 4 MB	NPU
ResNeXt101	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	3.937 ms	0 - 212 MB	NPU
ResNeXt101	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	1.578 ms	0 - 216 MB	NPU
ResNeXt101	ONNX	w8a8	Snapdragon® 8 Elite Mobile	1.892 ms	0 - 219 MB	NPU
ResNeXt101	ONNX	w8a8	Snapdragon® X2 Elite	1.397 ms	87 - 87 MB	NPU
ResNeXt101	ONNX	w8a8	Snapdragon® X Elite	3.139 ms	87 - 87 MB	NPU
ResNeXt101	ONNX	w8a8	Snapdragon® X Elite	3.139 ms	87 - 87 MB	NPU
ResNeXt101	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	2.164 ms	0 - 257 MB	NPU
ResNeXt101	ONNX	w8a8	Qualcomm® QCS6490	113.523 ms	7 - 49 MB	CPU
ResNeXt101	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	2.894 ms	0 - 100 MB	NPU
ResNeXt101	ONNX	w8a8	Qualcomm® QCS9075	3.148 ms	0 - 3 MB	NPU
ResNeXt101	ONNX	w8a8	Qualcomm® QCM6690	73.881 ms	14 - 30 MB	CPU
ResNeXt101	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.892 ms	0 - 219 MB	NPU
ResNeXt101	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	68.112 ms	0 - 12 MB	CPU
ResNeXt101	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	68.112 ms	0 - 12 MB	CPU
ResNeXt101	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	3.14 ms	1 - 191 MB	NPU
ResNeXt101	QNN_DLC	float	Snapdragon® 8 Elite Mobile	4.02 ms	1 - 180 MB	NPU
ResNeXt101	QNN_DLC	float	Snapdragon® X2 Elite	3.601 ms	1 - 1 MB	NPU
ResNeXt101	QNN_DLC	float	Snapdragon® X Elite	6.927 ms	1 - 1 MB	NPU
ResNeXt101	QNN_DLC	float	Snapdragon® X Elite	6.927 ms	1 - 1 MB	NPU
ResNeXt101	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	4.649 ms	0 - 366 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	35.994 ms	1 - 180 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	6.86 ms	1 - 3 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® SA8775P	10.273 ms	1 - 194 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® SA8775P	10.273 ms	1 - 194 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® SA8775P	10.273 ms	1 - 194 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® QCS9075	9.978 ms	3 - 5 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	9.991 ms	0 - 308 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® SA7255P	35.994 ms	1 - 180 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® SA8295P	10.908 ms	1 - 134 MB	NPU
ResNeXt101	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	4.02 ms	1 - 180 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	1.501 ms	0 - 214 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® 8 Elite Mobile	1.848 ms	0 - 209 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® X2 Elite	1.641 ms	0 - 0 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® X Elite	3.064 ms	0 - 0 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® X Elite	3.064 ms	0 - 0 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	2.145 ms	0 - 258 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® QCS6490	9.281 ms	0 - 2 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	6.612 ms	0 - 211 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	2.83 ms	0 - 232 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® SA8775P	3.501 ms	0 - 212 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® SA8775P	3.501 ms	0 - 212 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® SA8775P	3.501 ms	0 - 212 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® QCS9075	3.108 ms	0 - 2 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® QCM6690	30.741 ms	0 - 371 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	3.981 ms	0 - 258 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® SA7255P	6.612 ms	0 - 211 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® SA8295P	4.284 ms	0 - 214 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.848 ms	0 - 209 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	4.04 ms	0 - 243 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	4.04 ms	0 - 243 MB	NPU
ResNeXt101	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	3.117 ms	0 - 365 MB	NPU
ResNeXt101	TFLITE	float	Snapdragon® 8 Elite Mobile	3.958 ms	0 - 355 MB	NPU
ResNeXt101	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	4.632 ms	0 - 532 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® QCS8275 (Proxy)	35.998 ms	0 - 358 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® QCS8550 (Proxy)	6.9 ms	0 - 3 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® SA8775P	10.273 ms	0 - 358 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® SA8775P	10.273 ms	0 - 358 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® SA8775P	10.273 ms	0 - 358 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® QCS9075	10.026 ms	0 - 174 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® QCS8450 (Proxy)	10.074 ms	0 - 478 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® SA7255P	35.998 ms	0 - 358 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® SA8295P	10.972 ms	0 - 305 MB	NPU
ResNeXt101	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	3.958 ms	0 - 355 MB	NPU
ResNeXt101	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	1.466 ms	0 - 212 MB	NPU
ResNeXt101	TFLITE	w8a8	Snapdragon® 8 Elite Mobile	1.733 ms	0 - 209 MB	NPU
ResNeXt101	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	1.988 ms	0 - 249 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® QCS6490	8.956 ms	0 - 89 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	6.268 ms	0 - 214 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	2.718 ms	0 - 2 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® SA8775P	3.326 ms	0 - 215 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® SA8775P	3.326 ms	0 - 215 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® SA8775P	3.326 ms	0 - 215 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® QCS9075	2.945 ms	0 - 89 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® QCM6690	28.848 ms	0 - 375 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	3.719 ms	0 - 255 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® SA7255P	6.268 ms	0 - 214 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® SA8295P	4.09 ms	0 - 216 MB	NPU
ResNeXt101	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.733 ms	0 - 209 MB	NPU
ResNeXt101	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	3.927 ms	0 - 242 MB	NPU
ResNeXt101	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	3.927 ms	0 - 242 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.