---
library_name: pytorch
license: other
tags:
- android
pipeline_tag: image-classification

---

![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/shufflenet_v2/web-assets/model_demo.png)

# Shufflenet-v2: Optimized for Qualcomm Devices

ShufflenetV2 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 Shufflenet-v2 found [here](https://github.com/pytorch/vision/blob/main/torchvision/models/shufflenetv2.py).
This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/shufflenet_v2) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).

Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) 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](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/shufflenet_v2/releases/v0.51.0/shufflenet_v2-onnx-float.zip)
| ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/shufflenet_v2/releases/v0.51.0/shufflenet_v2-onnx-w8a8.zip)
| QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/shufflenet_v2/releases/v0.51.0/shufflenet_v2-qnn_dlc-float.zip)
| QNN_DLC | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/shufflenet_v2/releases/v0.51.0/shufflenet_v2-qnn_dlc-w8a8.zip)
| TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/shufflenet_v2/releases/v0.51.0/shufflenet_v2-tflite-float.zip)
| TFLITE | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/shufflenet_v2/releases/v0.51.0/shufflenet_v2-tflite-w8a8.zip)

For more device-specific assets and performance metrics, visit **[Shufflenet-v2 on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/shufflenet_v2)**.


### Option 2: Export with Custom Configurations

Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/shufflenet_v2) 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 [Shufflenet-v2 on GitHub](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/shufflenet_v2) for usage instructions.

## Model Details

**Model Type:** Model_use_case.image_classification

**Model Stats:**
- Model checkpoint: Imagenet
- Input resolution: 224x224
- Number of parameters: 1.37M
- Model size (float): 5.24 MB
- Model size (w8a8): 1.47 MB

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| Shufflenet-v2 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.426 ms | 0 - 32 MB | NPU
| Shufflenet-v2 | ONNX | float | Snapdragon® X2 Elite | 0.425 ms | 0 - 0 MB | NPU
| Shufflenet-v2 | ONNX | float | Snapdragon® X Elite | 0.966 ms | 2 - 2 MB | NPU
| Shufflenet-v2 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.51 ms | 0 - 41 MB | NPU
| Shufflenet-v2 | ONNX | float | Qualcomm® QCS8550 (Proxy) | 0.819 ms | 0 - 5 MB | NPU
| Shufflenet-v2 | ONNX | float | Qualcomm® QCS9075 | 0.995 ms | 1 - 3 MB | NPU
| Shufflenet-v2 | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.434 ms | 0 - 33 MB | NPU
| Shufflenet-v2 | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.34 ms | 0 - 31 MB | NPU
| Shufflenet-v2 | ONNX | w8a8 | Snapdragon® X2 Elite | 0.332 ms | 0 - 0 MB | NPU
| Shufflenet-v2 | ONNX | w8a8 | Snapdragon® X Elite | 0.709 ms | 0 - 0 MB | NPU
| Shufflenet-v2 | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.411 ms | 0 - 37 MB | NPU
| Shufflenet-v2 | ONNX | w8a8 | Qualcomm® QCS6490 | 3.231 ms | 4 - 7 MB | CPU
| Shufflenet-v2 | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.576 ms | 0 - 13 MB | NPU
| Shufflenet-v2 | ONNX | w8a8 | Qualcomm® QCS9075 | 0.686 ms | 0 - 3 MB | NPU
| Shufflenet-v2 | ONNX | w8a8 | Qualcomm® QCM6690 | 2.196 ms | 0 - 9 MB | CPU
| Shufflenet-v2 | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.356 ms | 0 - 32 MB | NPU
| Shufflenet-v2 | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.478 ms | 0 - 9 MB | CPU
| Shufflenet-v2 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.278 ms | 1 - 30 MB | NPU
| Shufflenet-v2 | QNN_DLC | float | Snapdragon® X2 Elite | 0.412 ms | 1 - 1 MB | NPU
| Shufflenet-v2 | QNN_DLC | float | Snapdragon® X Elite | 0.931 ms | 1 - 1 MB | NPU
| Shufflenet-v2 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 0.504 ms | 0 - 38 MB | NPU
| Shufflenet-v2 | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 1.725 ms | 1 - 26 MB | NPU
| Shufflenet-v2 | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 0.785 ms | 0 - 7 MB | NPU
| Shufflenet-v2 | QNN_DLC | float | Qualcomm® SA8775P | 0.998 ms | 1 - 29 MB | NPU
| Shufflenet-v2 | QNN_DLC | float | Qualcomm® QCS9075 | 0.897 ms | 3 - 5 MB | NPU
| Shufflenet-v2 | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 1.358 ms | 0 - 40 MB | NPU
| Shufflenet-v2 | QNN_DLC | float | Qualcomm® SA7255P | 1.725 ms | 1 - 26 MB | NPU
| Shufflenet-v2 | QNN_DLC | float | Qualcomm® SA8295P | 1.225 ms | 0 - 25 MB | NPU
| Shufflenet-v2 | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.362 ms | 1 - 31 MB | NPU
| Shufflenet-v2 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.204 ms | 0 - 26 MB | NPU
| Shufflenet-v2 | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.337 ms | 0 - 0 MB | NPU
| Shufflenet-v2 | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.588 ms | 0 - 0 MB | NPU
| Shufflenet-v2 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.334 ms | 0 - 31 MB | NPU
| Shufflenet-v2 | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 1.163 ms | 0 - 2 MB | NPU
| Shufflenet-v2 | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 1.061 ms | 0 - 23 MB | NPU
| Shufflenet-v2 | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.472 ms | 0 - 16 MB | NPU
| Shufflenet-v2 | QNN_DLC | w8a8 | Qualcomm® SA8775P | 0.648 ms | 0 - 25 MB | NPU
| Shufflenet-v2 | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 0.56 ms | 2 - 4 MB | NPU
| Shufflenet-v2 | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 1.336 ms | 0 - 23 MB | NPU
| Shufflenet-v2 | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.541 ms | 0 - 33 MB | NPU
| Shufflenet-v2 | QNN_DLC | w8a8 | Qualcomm® SA7255P | 1.061 ms | 0 - 23 MB | NPU
| Shufflenet-v2 | QNN_DLC | w8a8 | Qualcomm® SA8295P | 0.828 ms | 0 - 21 MB | NPU
| Shufflenet-v2 | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.264 ms | 0 - 22 MB | NPU
| Shufflenet-v2 | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.458 ms | 0 - 22 MB | NPU
| Shufflenet-v2 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.281 ms | 0 - 30 MB | NPU
| Shufflenet-v2 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 0.499 ms | 0 - 39 MB | NPU
| Shufflenet-v2 | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 1.736 ms | 0 - 27 MB | NPU
| Shufflenet-v2 | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 0.777 ms | 0 - 18 MB | NPU
| Shufflenet-v2 | TFLITE | float | Qualcomm® SA8775P | 1.02 ms | 0 - 29 MB | NPU
| Shufflenet-v2 | TFLITE | float | Qualcomm® QCS9075 | 0.887 ms | 0 - 6 MB | NPU
| Shufflenet-v2 | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 1.367 ms | 0 - 39 MB | NPU
| Shufflenet-v2 | TFLITE | float | Qualcomm® SA7255P | 1.736 ms | 0 - 27 MB | NPU
| Shufflenet-v2 | TFLITE | float | Qualcomm® SA8295P | 1.255 ms | 0 - 24 MB | NPU
| Shufflenet-v2 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.367 ms | 0 - 27 MB | NPU
| Shufflenet-v2 | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.252 ms | 0 - 27 MB | NPU
| Shufflenet-v2 | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.321 ms | 0 - 32 MB | NPU
| Shufflenet-v2 | TFLITE | w8a8 | Qualcomm® QCS6490 | 0.875 ms | 0 - 4 MB | NPU
| Shufflenet-v2 | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 1.038 ms | 0 - 23 MB | NPU
| Shufflenet-v2 | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.456 ms | 0 - 1 MB | NPU
| Shufflenet-v2 | TFLITE | w8a8 | Qualcomm® SA8775P | 0.651 ms | 0 - 26 MB | NPU
| Shufflenet-v2 | TFLITE | w8a8 | Qualcomm® QCS9075 | 0.568 ms | 0 - 3 MB | NPU
| Shufflenet-v2 | TFLITE | w8a8 | Qualcomm® QCM6690 | 1.061 ms | 0 - 23 MB | NPU
| Shufflenet-v2 | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.512 ms | 0 - 33 MB | NPU
| Shufflenet-v2 | TFLITE | w8a8 | Qualcomm® SA7255P | 1.038 ms | 0 - 23 MB | NPU
| Shufflenet-v2 | TFLITE | w8a8 | Qualcomm® SA8295P | 0.821 ms | 0 - 21 MB | NPU
| Shufflenet-v2 | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.273 ms | 0 - 24 MB | NPU
| Shufflenet-v2 | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.442 ms | 0 - 23 MB | NPU

## License
* The license for the original implementation of Shufflenet-v2 can be found
  [here](https://github.com/pytorch/vision/blob/main/LICENSE).

## References
* [ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design](https://arxiv.org/abs/1807.11164)
* [Source Model Implementation](https://github.com/pytorch/vision/blob/main/torchvision/models/shufflenetv2.py)

## Community
* Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
* For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).