Sequencer2D: Optimized for Qualcomm Devices

Sequencer2D is a vision transformer model that can classify images from the Imagenet dataset.

This is based on the implementation of Sequencer2D 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
TFLITE	float	Universal	QAIRT 2.45	Download

For more device-specific assets and performance metrics, visit Sequencer2D 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 Sequencer2D on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.image_classification

Model Stats:

• Model checkpoint: sequencer2d_s
• Input resolution: 224x224
• Number of parameters: 27.6M
• Model size (float): 106 MB
• Model size (w8a8): 69.1 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
Sequencer2D	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	7.882 ms	0 - 1367 MB	NPU
Sequencer2D	ONNX	float	Snapdragon® X2 Elite	9.777 ms	61 - 61 MB	NPU
Sequencer2D	ONNX	float	Snapdragon® X Elite	16.97 ms	59 - 59 MB	NPU
Sequencer2D	ONNX	float	Snapdragon® 8 Gen 3 Mobile	11.274 ms	1 - 2286 MB	NPU
Sequencer2D	ONNX	float	Qualcomm® QCS8550 (Proxy)	16.727 ms	0 - 82 MB	NPU
Sequencer2D	ONNX	float	Qualcomm® QCS9075	19.6 ms	0 - 4 MB	NPU
Sequencer2D	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	9.831 ms	1 - 1046 MB	NPU
Sequencer2D	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	8.08 ms	0 - 730 MB	NPU
Sequencer2D	ONNX	w8a16	Snapdragon® X2 Elite	9.939 ms	45 - 45 MB	NPU
Sequencer2D	ONNX	w8a16	Snapdragon® X Elite	18.164 ms	43 - 43 MB	NPU
Sequencer2D	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	11.437 ms	0 - 869 MB	NPU
Sequencer2D	ONNX	w8a16	Qualcomm® QCS6490	612.837 ms	46 - 52 MB	CPU
Sequencer2D	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	17.491 ms	0 - 71 MB	NPU
Sequencer2D	ONNX	w8a16	Qualcomm® QCS9075	20.543 ms	0 - 3 MB	NPU
Sequencer2D	ONNX	w8a16	Qualcomm® QCM6690	310.143 ms	47 - 63 MB	CPU
Sequencer2D	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	10.125 ms	0 - 705 MB	NPU
Sequencer2D	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	298.712 ms	47 - 65 MB	CPU
Sequencer2D	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	9.208 ms	0 - 855 MB	NPU
Sequencer2D	QNN_DLC	float	Snapdragon® X2 Elite	14.418 ms	1 - 1 MB	NPU
Sequencer2D	QNN_DLC	float	Snapdragon® X Elite	21.702 ms	1 - 1 MB	NPU
Sequencer2D	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	14.436 ms	0 - 2148 MB	NPU
Sequencer2D	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	52.203 ms	1 - 973 MB	NPU
Sequencer2D	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	20.907 ms	1 - 424 MB	NPU
Sequencer2D	QNN_DLC	float	Qualcomm® QCS9075	23.027 ms	1 - 3 MB	NPU
Sequencer2D	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	26.331 ms	0 - 797 MB	NPU
Sequencer2D	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	12.196 ms	1 - 1047 MB	NPU
Sequencer2D	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	7.192 ms	0 - 756 MB	NPU
Sequencer2D	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	11.925 ms	0 - 981 MB	NPU
Sequencer2D	TFLITE	float	Qualcomm® QCS8275 (Proxy)	37.327 ms	0 - 796 MB	NPU
Sequencer2D	TFLITE	float	Qualcomm® QCS8550 (Proxy)	17.45 ms	0 - 18 MB	NPU
Sequencer2D	TFLITE	float	Qualcomm® QCS9075	20.36 ms	0 - 75 MB	NPU
Sequencer2D	TFLITE	float	Qualcomm® QCS8450 (Proxy)	21.317 ms	0 - 726 MB	NPU
Sequencer2D	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	9.395 ms	0 - 784 MB	NPU

License

• The license for the original implementation of Sequencer2D can be found here.

References

• Sequencer: Deep LSTM for Image Classification
• 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.