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	3.047 ms	1 - 203 MB	NPU
ResNeXt101	ONNX	float	Snapdragon® X2 Elite	3.095 ms	173 - 173 MB	NPU
ResNeXt101	ONNX	float	Snapdragon® X Elite	6.716 ms	172 - 172 MB	NPU
ResNeXt101	ONNX	float	Snapdragon® 8 Gen 3 Mobile	4.525 ms	0 - 386 MB	NPU
ResNeXt101	ONNX	float	Qualcomm® QCS8550 (Proxy)	6.736 ms	0 - 196 MB	NPU
ResNeXt101	ONNX	float	Qualcomm® QCS9075	9.763 ms	0 - 4 MB	NPU
ResNeXt101	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	3.935 ms	0 - 212 MB	NPU
ResNeXt101	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	1.573 ms	0 - 217 MB	NPU
ResNeXt101	ONNX	w8a8	Snapdragon® X2 Elite	1.388 ms	87 - 87 MB	NPU
ResNeXt101	ONNX	w8a8	Snapdragon® X Elite	3.128 ms	87 - 87 MB	NPU
ResNeXt101	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	2.148 ms	0 - 254 MB	NPU
ResNeXt101	ONNX	w8a8	Qualcomm® QCS6490	112.859 ms	12 - 44 MB	CPU
ResNeXt101	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	2.993 ms	0 - 99 MB	NPU
ResNeXt101	ONNX	w8a8	Qualcomm® QCS9075	3.133 ms	0 - 3 MB	NPU
ResNeXt101	ONNX	w8a8	Qualcomm® QCM6690	73.566 ms	1 - 13 MB	CPU
ResNeXt101	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.891 ms	0 - 218 MB	NPU
ResNeXt101	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	69.187 ms	0 - 12 MB	CPU
ResNeXt101	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	3.146 ms	0 - 188 MB	NPU
ResNeXt101	QNN_DLC	float	Snapdragon® X2 Elite	3.661 ms	1 - 1 MB	NPU
ResNeXt101	QNN_DLC	float	Snapdragon® X Elite	6.916 ms	1 - 1 MB	NPU
ResNeXt101	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	4.628 ms	0 - 364 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	35.977 ms	1 - 180 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	6.75 ms	1 - 346 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® SA8775P	10.288 ms	1 - 194 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® QCS9075	9.967 ms	3 - 5 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	10.004 ms	0 - 309 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® SA7255P	35.977 ms	1 - 180 MB	NPU
ResNeXt101	QNN_DLC	float	Qualcomm® SA8295P	10.928 ms	1 - 134 MB	NPU
ResNeXt101	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	4.043 ms	1 - 179 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	1.503 ms	0 - 215 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® X2 Elite	1.588 ms	0 - 0 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® X Elite	3.063 ms	0 - 0 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	2.104 ms	0 - 256 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® QCS6490	9.332 ms	2 - 4 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	6.543 ms	0 - 210 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	2.804 ms	0 - 138 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® SA8775P	3.49 ms	0 - 212 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® QCS9075	3.096 ms	2 - 4 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® QCM6690	28.149 ms	0 - 371 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	3.98 ms	0 - 259 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® SA7255P	6.543 ms	0 - 210 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Qualcomm® SA8295P	4.252 ms	0 - 214 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.825 ms	0 - 208 MB	NPU
ResNeXt101	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	4.054 ms	0 - 242 MB	NPU
ResNeXt101	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	3.096 ms	0 - 364 MB	NPU
ResNeXt101	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	4.586 ms	0 - 528 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® QCS8275 (Proxy)	35.986 ms	0 - 359 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® QCS8550 (Proxy)	6.842 ms	0 - 3 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® SA8775P	10.224 ms	0 - 358 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® QCS9075	10.023 ms	0 - 174 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® QCS8450 (Proxy)	9.861 ms	0 - 477 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® SA7255P	35.986 ms	0 - 359 MB	NPU
ResNeXt101	TFLITE	float	Qualcomm® SA8295P	10.973 ms	0 - 304 MB	NPU
ResNeXt101	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	3.986 ms	0 - 360 MB	NPU
ResNeXt101	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	1.453 ms	0 - 213 MB	NPU
ResNeXt101	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	1.968 ms	0 - 257 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® QCS6490	8.898 ms	0 - 88 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	6.265 ms	0 - 213 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	2.736 ms	0 - 4 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® SA8775P	3.313 ms	0 - 214 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® QCS9075	2.92 ms	0 - 89 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® QCM6690	26.78 ms	0 - 375 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	3.806 ms	0 - 255 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® SA7255P	6.265 ms	0 - 213 MB	NPU
ResNeXt101	TFLITE	w8a8	Qualcomm® SA8295P	4.094 ms	0 - 216 MB	NPU
ResNeXt101	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.736 ms	0 - 212 MB	NPU
ResNeXt101	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	3.94 ms	0 - 243 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.