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 2.42, ONNX Runtime 1.24.1	Download
ONNX	w8a16	Universal	QAIRT 2.42, ONNX Runtime 1.24.1	Download
QNN_DLC	float	Universal	QAIRT 2.43	Download
QNN_DLC	w8a16	Universal	QAIRT 2.43	Download
TFLITE	float	Universal	QAIRT 2.43, TFLite 2.17.0	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® X Elite	7.505 ms	175 - 175 MB	NPU
ConvNext-Base	ONNX	float	Snapdragon® 8 Gen 3 Mobile	5.3 ms	1 - 352 MB	NPU
ConvNext-Base	ONNX	float	Qualcomm® QCS8550 (Proxy)	7.15 ms	0 - 195 MB	NPU
ConvNext-Base	ONNX	float	Qualcomm® QCS9075	11.469 ms	0 - 4 MB	NPU
ConvNext-Base	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	4.118 ms	0 - 284 MB	NPU
ConvNext-Base	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	3.159 ms	1 - 285 MB	NPU
ConvNext-Base	ONNX	float	Snapdragon® X2 Elite	3.54 ms	176 - 176 MB	NPU
ConvNext-Base	ONNX	w8a16	Snapdragon® X Elite	6.457 ms	90 - 90 MB	NPU
ConvNext-Base	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	4.39 ms	0 - 270 MB	NPU
ConvNext-Base	ONNX	w8a16	Qualcomm® QCS6490	1091.858 ms	32 - 63 MB	CPU
ConvNext-Base	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	6.185 ms	0 - 359 MB	NPU
ConvNext-Base	ONNX	w8a16	Qualcomm® QCS9075	5.895 ms	0 - 3 MB	NPU
ConvNext-Base	ONNX	w8a16	Qualcomm® QCM6690	630.675 ms	43 - 55 MB	CPU
ConvNext-Base	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	3.205 ms	0 - 208 MB	NPU
ConvNext-Base	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	605.16 ms	74 - 89 MB	CPU
ConvNext-Base	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	2.596 ms	0 - 223 MB	NPU
ConvNext-Base	ONNX	w8a16	Snapdragon® X2 Elite	2.779 ms	90 - 90 MB	NPU
ConvNext-Base	QNN_DLC	float	Snapdragon® X Elite	8.57 ms	1 - 1 MB	NPU
ConvNext-Base	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	6.002 ms	0 - 347 MB	NPU
ConvNext-Base	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	42.209 ms	1 - 278 MB	NPU
ConvNext-Base	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	8.247 ms	1 - 448 MB	NPU
ConvNext-Base	QNN_DLC	float	Qualcomm® QCS9075	12.123 ms	1 - 3 MB	NPU
ConvNext-Base	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	20.649 ms	0 - 336 MB	NPU
ConvNext-Base	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	4.655 ms	1 - 281 MB	NPU
ConvNext-Base	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	3.556 ms	0 - 282 MB	NPU
ConvNext-Base	QNN_DLC	float	Snapdragon® X2 Elite	4.311 ms	1 - 1 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Snapdragon® X Elite	6.27 ms	0 - 0 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	4.071 ms	0 - 247 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Qualcomm® QCS6490	23.725 ms	2 - 4 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Qualcomm® QCS8275 (Proxy)	14.607 ms	0 - 198 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Qualcomm® QCS8550 (Proxy)	5.941 ms	0 - 2 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Qualcomm® QCS9075	6.136 ms	0 - 2 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Qualcomm® QCM6690	76.067 ms	0 - 394 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Qualcomm® QCS8450 (Proxy)	9.092 ms	0 - 245 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	3.28 ms	0 - 189 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	7.763 ms	0 - 249 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	2.509 ms	0 - 201 MB	NPU
ConvNext-Base	QNN_DLC	w8a16	Snapdragon® X2 Elite	3.056 ms	0 - 0 MB	NPU
ConvNext-Base	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	5.443 ms	0 - 343 MB	NPU
ConvNext-Base	TFLITE	float	Qualcomm® QCS8275 (Proxy)	40.924 ms	0 - 274 MB	NPU
ConvNext-Base	TFLITE	float	Qualcomm® QCS8550 (Proxy)	7.25 ms	0 - 2 MB	NPU
ConvNext-Base	TFLITE	float	Qualcomm® QCS9075	11.45 ms	0 - 177 MB	NPU
ConvNext-Base	TFLITE	float	Qualcomm® QCS8450 (Proxy)	19.581 ms	0 - 331 MB	NPU
ConvNext-Base	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	4.101 ms	0 - 274 MB	NPU
ConvNext-Base	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	3.157 ms	0 - 278 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.