ConvNext-Tiny: Optimized for Qualcomm Devices

ConvNextTiny 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-Tiny 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	w8a16	Universal	QAIRT 2.42, ONNX Runtime 1.24.3	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-Tiny 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-Tiny 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: 28.6M
• Model size (float): 109 MB
• Model size (w8a16): 28.9 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
ConvNext-Tiny	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	1.281 ms	0 - 127 MB	NPU
ConvNext-Tiny	ONNX	float	Snapdragon® X2 Elite	1.349 ms	57 - 57 MB	NPU
ConvNext-Tiny	ONNX	float	Snapdragon® X Elite	2.9 ms	56 - 56 MB	NPU
ConvNext-Tiny	ONNX	float	Snapdragon® 8 Gen 3 Mobile	2.038 ms	0 - 171 MB	NPU
ConvNext-Tiny	ONNX	float	Qualcomm® QCS8550 (Proxy)	2.735 ms	1 - 5 MB	NPU
ConvNext-Tiny	ONNX	float	Qualcomm® QCS9075	3.942 ms	0 - 4 MB	NPU
ConvNext-Tiny	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	1.555 ms	0 - 127 MB	NPU
ConvNext-Tiny	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	1.098 ms	0 - 115 MB	NPU
ConvNext-Tiny	ONNX	w8a16	Snapdragon® X2 Elite	1.202 ms	29 - 29 MB	NPU
ConvNext-Tiny	ONNX	w8a16	Snapdragon® X Elite	2.835 ms	29 - 29 MB	NPU
ConvNext-Tiny	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	1.795 ms	0 - 141 MB	NPU
ConvNext-Tiny	ONNX	w8a16	Qualcomm® QCS6490	408.527 ms	50 - 64 MB	CPU
ConvNext-Tiny	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	2.562 ms	0 - 35 MB	NPU
ConvNext-Tiny	ONNX	w8a16	Qualcomm® QCS9075	2.669 ms	0 - 3 MB	NPU
ConvNext-Tiny	ONNX	w8a16	Qualcomm® QCM6690	208.906 ms	60 - 73 MB	CPU
ConvNext-Tiny	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	1.377 ms	0 - 110 MB	NPU
ConvNext-Tiny	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	199.714 ms	57 - 72 MB	CPU
ConvNext-Tiny	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	1.604 ms	1 - 79 MB	NPU
ConvNext-Tiny	QNN_DLC	float	Snapdragon® X2 Elite	2.008 ms	1 - 1 MB	NPU
ConvNext-Tiny	QNN_DLC	float	Snapdragon® X Elite	3.944 ms	1 - 1 MB	NPU
ConvNext-Tiny	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	2.65 ms	0 - 127 MB	NPU
ConvNext-Tiny	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	15.285 ms	1 - 76 MB	NPU
ConvNext-Tiny	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	3.651 ms	1 - 2 MB	NPU
ConvNext-Tiny	QNN_DLC	float	Qualcomm® SA8775P	5.072 ms	1 - 76 MB	NPU
ConvNext-Tiny	QNN_DLC	float	Qualcomm® QCS9075	4.855 ms	1 - 3 MB	NPU
ConvNext-Tiny	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	9.687 ms	0 - 128 MB	NPU
ConvNext-Tiny	QNN_DLC	float	Qualcomm® SA7255P	15.285 ms	1 - 76 MB	NPU
ConvNext-Tiny	QNN_DLC	float	Qualcomm® SA8295P	8.938 ms	1 - 78 MB	NPU
ConvNext-Tiny	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	2.026 ms	1 - 77 MB	NPU
ConvNext-Tiny	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	1.276 ms	0 - 101 MB	NPU
ConvNext-Tiny	QNN_DLC	w8a16	Snapdragon® X2 Elite	1.589 ms	0 - 0 MB	NPU
ConvNext-Tiny	QNN_DLC	w8a16	Snapdragon® X Elite	3.35 ms	0 - 0 MB	NPU
ConvNext-Tiny	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	2.181 ms	0 - 122 MB	NPU
ConvNext-Tiny	QNN_DLC	w8a16	Qualcomm® QCS6490	9.054 ms	2 - 4 MB	NPU
ConvNext-Tiny	QNN_DLC	w8a16	Qualcomm® QCS8275 (Proxy)	6.835 ms	0 - 96 MB	NPU
ConvNext-Tiny	QNN_DLC	w8a16	Qualcomm® QCS8550 (Proxy)	3.108 ms	0 - 2 MB	NPU
ConvNext-Tiny	QNN_DLC	w8a16	Qualcomm® SA8775P	3.522 ms	0 - 97 MB	NPU
ConvNext-Tiny	QNN_DLC	w8a16	Qualcomm® QCS9075	3.327 ms	2 - 4 MB	NPU
ConvNext-Tiny	QNN_DLC	w8a16	Qualcomm® QCM6690	22.296 ms	0 - 249 MB	NPU
ConvNext-Tiny	QNN_DLC	w8a16	Qualcomm® QCS8450 (Proxy)	4.225 ms	0 - 121 MB	NPU
ConvNext-Tiny	QNN_DLC	w8a16	Qualcomm® SA7255P	6.835 ms	0 - 96 MB	NPU
ConvNext-Tiny	QNN_DLC	w8a16	Qualcomm® SA8295P	4.737 ms	0 - 93 MB	NPU
ConvNext-Tiny	QNN_DLC	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	1.589 ms	0 - 99 MB	NPU
ConvNext-Tiny	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	3.41 ms	0 - 110 MB	NPU
ConvNext-Tiny	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	1.309 ms	0 - 77 MB	NPU
ConvNext-Tiny	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	2.145 ms	0 - 127 MB	NPU
ConvNext-Tiny	TFLITE	float	Qualcomm® QCS8275 (Proxy)	14.043 ms	0 - 73 MB	NPU
ConvNext-Tiny	TFLITE	float	Qualcomm® QCS8550 (Proxy)	2.837 ms	0 - 2 MB	NPU
ConvNext-Tiny	TFLITE	float	Qualcomm® SA8775P	4.253 ms	0 - 75 MB	NPU
ConvNext-Tiny	TFLITE	float	Qualcomm® QCS9075	4.064 ms	0 - 59 MB	NPU
ConvNext-Tiny	TFLITE	float	Qualcomm® QCS8450 (Proxy)	8.884 ms	0 - 123 MB	NPU
ConvNext-Tiny	TFLITE	float	Qualcomm® SA7255P	14.043 ms	0 - 73 MB	NPU
ConvNext-Tiny	TFLITE	float	Qualcomm® SA8295P	7.839 ms	0 - 71 MB	NPU
ConvNext-Tiny	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	1.588 ms	0 - 72 MB	NPU

License

• The license for the original implementation of ConvNext-Tiny 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.