SqueezeNet-1.1: Optimized for Qualcomm Devices
SqueezeNet 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 SqueezeNet-1.1 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 | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.3 | Download |
| QNN_DLC | float | Universal | QAIRT 2.45 | Download |
| QNN_DLC | w8a8 | Universal | QAIRT 2.45 | Download |
| TFLITE | float | Universal | QAIRT 2.45 | Download |
| TFLITE | w8a8 | Universal | QAIRT 2.45 | Download |
For more device-specific assets and performance metrics, visit SqueezeNet-1.1 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 SqueezeNet-1.1 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: 1.24M
- Model size (float): 4.73 MB
- Model size (w8a8): 1.30 MB
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| SqueezeNet-1.1 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.177 ms | 0 - 23 MB | NPU |
| SqueezeNet-1.1 | ONNX | float | Snapdragon® 8 Elite Mobile | 0.218 ms | 0 - 24 MB | NPU |
| SqueezeNet-1.1 | ONNX | float | Snapdragon® X2 Elite | 0.183 ms | 0 - 0 MB | NPU |
| SqueezeNet-1.1 | ONNX | float | Snapdragon® X Elite | 0.507 ms | 2 - 2 MB | NPU |
| SqueezeNet-1.1 | ONNX | float | Snapdragon® X Elite | 0.507 ms | 2 - 2 MB | NPU |
| SqueezeNet-1.1 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.277 ms | 0 - 30 MB | NPU |
| SqueezeNet-1.1 | ONNX | float | Qualcomm® QCS8550 (Proxy) | 0.382 ms | 0 - 4 MB | NPU |
| SqueezeNet-1.1 | ONNX | float | Qualcomm® QCS9075 | 0.613 ms | 1 - 3 MB | NPU |
| SqueezeNet-1.1 | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.218 ms | 0 - 24 MB | NPU |
| SqueezeNet-1.1 | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.216 ms | 0 - 24 MB | NPU |
| SqueezeNet-1.1 | ONNX | w8a8 | Snapdragon® 8 Elite Mobile | 0.231 ms | 0 - 21 MB | NPU |
| SqueezeNet-1.1 | ONNX | w8a8 | Snapdragon® X2 Elite | 0.204 ms | 0 - 0 MB | NPU |
| SqueezeNet-1.1 | ONNX | w8a8 | Snapdragon® X Elite | 0.488 ms | 0 - 0 MB | NPU |
| SqueezeNet-1.1 | ONNX | w8a8 | Snapdragon® X Elite | 0.488 ms | 0 - 0 MB | NPU |
| SqueezeNet-1.1 | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.275 ms | 0 - 31 MB | NPU |
| SqueezeNet-1.1 | ONNX | w8a8 | Qualcomm® QCS6490 | 4.046 ms | 5 - 9 MB | CPU |
| SqueezeNet-1.1 | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.379 ms | 0 - 2 MB | NPU |
| SqueezeNet-1.1 | ONNX | w8a8 | Qualcomm® QCS9075 | 0.499 ms | 0 - 3 MB | NPU |
| SqueezeNet-1.1 | ONNX | w8a8 | Qualcomm® QCM6690 | 3.007 ms | 0 - 7 MB | CPU |
| SqueezeNet-1.1 | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.231 ms | 0 - 21 MB | NPU |
| SqueezeNet-1.1 | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 2.195 ms | 0 - 8 MB | CPU |
| SqueezeNet-1.1 | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 2.195 ms | 0 - 8 MB | CPU |
| SqueezeNet-1.1 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.24 ms | 1 - 25 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 0.326 ms | 1 - 21 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | float | Snapdragon® X2 Elite | 0.341 ms | 1 - 1 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | float | Snapdragon® X Elite | 0.806 ms | 1 - 1 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | float | Snapdragon® X Elite | 0.806 ms | 1 - 1 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 0.451 ms | 0 - 31 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 2.052 ms | 0 - 21 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 0.654 ms | 1 - 7 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | float | Qualcomm® SA8775P | 0.932 ms | 1 - 23 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | float | Qualcomm® SA8775P | 0.932 ms | 1 - 23 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | float | Qualcomm® SA8775P | 0.932 ms | 1 - 23 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | float | Qualcomm® QCS9075 | 0.866 ms | 1 - 3 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 1.272 ms | 0 - 33 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | float | Qualcomm® SA7255P | 2.052 ms | 0 - 21 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | float | Qualcomm® SA8295P | 1.135 ms | 0 - 19 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.326 ms | 1 - 21 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.147 ms | 0 - 23 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Mobile | 0.183 ms | 0 - 20 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.242 ms | 0 - 0 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.49 ms | 0 - 0 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.49 ms | 0 - 0 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.268 ms | 0 - 30 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 1.047 ms | 0 - 2 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 0.921 ms | 0 - 21 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.363 ms | 0 - 1 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Qualcomm® SA8775P | 0.536 ms | 0 - 23 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Qualcomm® SA8775P | 0.536 ms | 0 - 23 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Qualcomm® SA8775P | 0.536 ms | 0 - 23 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 0.456 ms | 0 - 2 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 1.435 ms | 0 - 20 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.434 ms | 0 - 31 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Qualcomm® SA7255P | 0.921 ms | 0 - 21 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Qualcomm® SA8295P | 0.72 ms | 0 - 19 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.183 ms | 0 - 20 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.371 ms | 0 - 20 MB | NPU |
| SqueezeNet-1.1 | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.371 ms | 0 - 20 MB | NPU |
| SqueezeNet-1.1 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.237 ms | 0 - 27 MB | NPU |
| SqueezeNet-1.1 | TFLITE | float | Snapdragon® 8 Elite Mobile | 0.333 ms | 0 - 24 MB | NPU |
| SqueezeNet-1.1 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 0.445 ms | 0 - 33 MB | NPU |
| SqueezeNet-1.1 | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 2.103 ms | 0 - 24 MB | NPU |
| SqueezeNet-1.1 | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 0.649 ms | 0 - 2 MB | NPU |
| SqueezeNet-1.1 | TFLITE | float | Qualcomm® SA8775P | 0.968 ms | 0 - 25 MB | NPU |
| SqueezeNet-1.1 | TFLITE | float | Qualcomm® SA8775P | 0.968 ms | 0 - 25 MB | NPU |
| SqueezeNet-1.1 | TFLITE | float | Qualcomm® SA8775P | 0.968 ms | 0 - 25 MB | NPU |
| SqueezeNet-1.1 | TFLITE | float | Qualcomm® QCS9075 | 0.87 ms | 0 - 5 MB | NPU |
| SqueezeNet-1.1 | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 1.266 ms | 0 - 34 MB | NPU |
| SqueezeNet-1.1 | TFLITE | float | Qualcomm® SA7255P | 2.103 ms | 0 - 24 MB | NPU |
| SqueezeNet-1.1 | TFLITE | float | Qualcomm® SA8295P | 1.161 ms | 0 - 21 MB | NPU |
| SqueezeNet-1.1 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.333 ms | 0 - 24 MB | NPU |
| SqueezeNet-1.1 | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.091 ms | 0 - 22 MB | NPU |
| SqueezeNet-1.1 | TFLITE | w8a8 | Snapdragon® 8 Elite Mobile | 0.116 ms | 0 - 24 MB | NPU |
| SqueezeNet-1.1 | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.144 ms | 0 - 29 MB | NPU |
| SqueezeNet-1.1 | TFLITE | w8a8 | Qualcomm® QCS6490 | 0.55 ms | 0 - 3 MB | NPU |
| SqueezeNet-1.1 | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 0.618 ms | 0 - 21 MB | NPU |
| SqueezeNet-1.1 | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.197 ms | 0 - 1 MB | NPU |
| SqueezeNet-1.1 | TFLITE | w8a8 | Qualcomm® SA8775P | 0.373 ms | 0 - 22 MB | NPU |
| SqueezeNet-1.1 | TFLITE | w8a8 | Qualcomm® SA8775P | 0.373 ms | 0 - 22 MB | NPU |
| SqueezeNet-1.1 | TFLITE | w8a8 | Qualcomm® SA8775P | 0.373 ms | 0 - 22 MB | NPU |
| SqueezeNet-1.1 | TFLITE | w8a8 | Qualcomm® QCS9075 | 0.274 ms | 0 - 3 MB | NPU |
| SqueezeNet-1.1 | TFLITE | w8a8 | Qualcomm® QCM6690 | 0.927 ms | 0 - 20 MB | NPU |
| SqueezeNet-1.1 | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.251 ms | 0 - 31 MB | NPU |
| SqueezeNet-1.1 | TFLITE | w8a8 | Qualcomm® SA7255P | 0.618 ms | 0 - 21 MB | NPU |
| SqueezeNet-1.1 | TFLITE | w8a8 | Qualcomm® SA8295P | 0.508 ms | 0 - 17 MB | NPU |
| SqueezeNet-1.1 | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.116 ms | 0 - 24 MB | NPU |
| SqueezeNet-1.1 | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.204 ms | 0 - 19 MB | NPU |
| SqueezeNet-1.1 | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.204 ms | 0 - 19 MB | NPU |
License
- The license for the original implementation of SqueezeNet-1.1 can be found here.
References
- SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
- 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.
