ConvNext-Base: Optimized for Qualcomm Devices

ConvNextBase 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 ConvNext-Base 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 2.45, ONNX Runtime 1.25.0	Download
ONNX	w8a16	Universal	QAIRT 2.45, ONNX Runtime 1.25.0	Download
QNN_DLC	float	Universal	QAIRT 2.45	Download
QNN_DLC	w8a16	Universal	QAIRT 2.45	Download
TFLITE	float	Universal	QAIRT 2.45	Download

For more device-specific assets and performance metrics, visit ConvNext-Base 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 ConvNext-Base 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: 88.6M
• Model size (float): 338 MB
• Model size (w8a16): 88.7 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
ConvNext-Base	ONNX	float	Snapdragon® X2 Elite	3.484 ms	2 - 2 MB	NPU
ConvNext-Base	ONNX	float	Snapdragon® X Elite	7.231 ms	177 - 177 MB	NPU
ConvNext-Base	ONNX	float	Snapdragon® 8 Gen 3 Mobile	5.337 ms	1 - 312 MB	NPU
ConvNext-Base	ONNX	float	Snapdragon® 8 Gen 1 Mobile	19.266 ms	1 - 300 MB	NPU
ConvNext-Base	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	7.157 ms	1 - 4 MB	NPU
ConvNext-Base	ONNX	float	Qualcomm® QCS8450	19.266 ms	1 - 300 MB	NPU
ConvNext-Base	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	10.779 ms	0 - 4 MB	NPU
ConvNext-Base	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	3.204 ms	0 - 183 MB	NPU
ConvNext-Base	ONNX	float	Snapdragon® 8 Elite Mobile	4.131 ms	0 - 183 MB	NPU
ConvNext-Base	ONNX	float	Qualcomm® Dragonwing™ Q-8750	4.131 ms	0 - 183 MB	NPU
ConvNext-Base	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	7.231 ms	177 - 177 MB	NPU
ConvNext-Base	ONNX	w8a16	Snapdragon® X2 Elite	2.385 ms	1 - 1 MB	NPU
ConvNext-Base	ONNX	w8a16	Snapdragon® X Elite	4.974 ms	91 - 91 MB	NPU
ConvNext-Base	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	3.415 ms	0 - 263 MB	NPU
ConvNext-Base	ONNX	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	4.846 ms	0 - 457 MB	NPU
ConvNext-Base	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-9075	4.877 ms	0 - 3 MB	NPU
ConvNext-Base	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	6.901 ms	0 - 261 MB	NPU
ConvNext-Base	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	2.169 ms	0 - 227 MB	NPU
ConvNext-Base	ONNX	w8a16	Qualcomm® Dragonwing™ Q-6690	58.306 ms	0 - 404 MB	NPU
ConvNext-Base	ONNX	w8a16	Snapdragon® 8 Elite Mobile	2.765 ms	0 - 206 MB	NPU
ConvNext-Base	ONNX	w8a16	Qualcomm® Dragonwing™ Q-7790	6.901 ms	0 - 261 MB	NPU
ConvNext-Base	ONNX	w8a16	Qualcomm® Dragonwing™ Q-8750	2.765 ms	0 - 206 MB	NPU
ConvNext-Base	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-X7181	4.974 ms	91 - 91 MB	NPU
ConvNext-Base	QNN_DLC	float	Snapdragon® X2 Elite	4.2 ms	1 - 1 MB	NPU
ConvNext-Base	QNN_DLC	float	Snapdragon® X Elite	8.397 ms	1 - 1 MB	NPU
ConvNext-Base	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	5.818 ms	0 - 305 MB	NPU
ConvNext-Base	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	20.395 ms	1 - 297 MB	NPU
ConvNext-Base	QNN_DLC	float	Qualcomm® QCS8275	41.911 ms	1 - 180 MB	NPU
ConvNext-Base	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	7.952 ms	1 - 2 MB	NPU
ConvNext-Base	QNN_DLC	float	Qualcomm® SA8775P	11.803 ms	1 - 181 MB	NPU
ConvNext-Base	QNN_DLC	float	Qualcomm® SA8650P	11.803 ms	1 - 181 MB	NPU
ConvNext-Base	QNN_DLC	float	Qualcomm® SA8255P	11.803 ms	1 - 181 MB	NPU
ConvNext-Base	QNN_DLC	float	Qualcomm® QCS8450	20.395 ms	1 - 297 MB	NPU
ConvNext-Base	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	12.069 ms	1 - 3 MB	NPU
ConvNext-Base	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	3.554 ms	1 - 185 MB	NPU
ConvNext-Base	QNN_DLC	float	Qualcomm® SA7255P	41.911 ms	1 - 180 MB	NPU
ConvNext-Base	QNN_DLC	float	Snapdragon® 8 Elite Mobile	4.53 ms	1 - 181 MB	NPU
ConvNext-Base	QNN_DLC	float	Qualcomm® SA8295P	19.651 ms	1 - 171 MB	NPU
ConvNext-Base	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	4.53 ms	1 - 181 MB	NPU
ConvNext-Base	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	8.397 ms	1 - 1 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Snapdragon® X2 Elite	3.079 ms	0 - 0 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Snapdragon® X Elite	6.316 ms	0 - 0 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	4.094 ms	0 - 251 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Qualcomm® QCS8275	14.612 ms	0 - 203 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	5.901 ms	0 - 2 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Qualcomm® SA8775P	6.2 ms	0 - 204 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Qualcomm® SA8650P	6.2 ms	0 - 204 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Qualcomm® SA8255P	6.2 ms	0 - 204 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-9075	6.843 ms	0 - 2 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	7.807 ms	0 - 254 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	2.559 ms	0 - 216 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-6690	60.79 ms	0 - 402 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Qualcomm® SA7255P	14.612 ms	0 - 203 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Snapdragon® 8 Elite Mobile	3.265 ms	0 - 195 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-7790	7.807 ms	0 - 254 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-8750	3.265 ms	0 - 195 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-X7181	6.316 ms	0 - 0 MB	NPU
ConvNext-Base	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	5.447 ms	0 - 300 MB	NPU
ConvNext-Base	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	19.624 ms	0 - 291 MB	NPU
ConvNext-Base	TFLITE	float	Qualcomm® QCS8275	40.966 ms	0 - 175 MB	NPU
ConvNext-Base	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	7.214 ms	0 - 3 MB	NPU
ConvNext-Base	TFLITE	float	Qualcomm® SA8775P	11.081 ms	0 - 176 MB	NPU
ConvNext-Base	TFLITE	float	Qualcomm® SA8650P	11.081 ms	0 - 176 MB	NPU
ConvNext-Base	TFLITE	float	Qualcomm® SA8255P	11.081 ms	0 - 176 MB	NPU
ConvNext-Base	TFLITE	float	Qualcomm® QCS8450	19.624 ms	0 - 291 MB	NPU
ConvNext-Base	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	10.817 ms	0 - 177 MB	NPU
ConvNext-Base	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	3.169 ms	0 - 182 MB	NPU
ConvNext-Base	TFLITE	float	Qualcomm® SA7255P	40.966 ms	0 - 175 MB	NPU
ConvNext-Base	TFLITE	float	Snapdragon® 8 Elite Mobile	4.107 ms	0 - 179 MB	NPU
ConvNext-Base	TFLITE	float	Qualcomm® SA8295P	18.793 ms	0 - 161 MB	NPU
ConvNext-Base	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	4.107 ms	0 - 179 MB	NPU

License

• The license for the original implementation of ConvNext-Base can be found here.

References

• A ConvNet for the 2020s
• 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.