A ConvNet for the 2020s
Paper • 2201.03545 • Published • 2
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 1.24.3 | 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-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® 8 Elite Gen 5 Mobile | 3.161 ms | 1 - 286 MB | NPU |
| ConvNext-Base | ONNX | float | Snapdragon® 8 Elite Mobile | 4.134 ms | 0 - 286 MB | NPU |
| ConvNext-Base | ONNX | float | Snapdragon® X2 Elite | 3.528 ms | 176 - 176 MB | NPU |
| ConvNext-Base | ONNX | float | Snapdragon® X Elite | 7.518 ms | 175 - 175 MB | NPU |
| ConvNext-Base | ONNX | float | Snapdragon® X Elite | 7.518 ms | 175 - 175 MB | NPU |
| ConvNext-Base | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 5.299 ms | 1 - 352 MB | NPU |
| ConvNext-Base | ONNX | float | Qualcomm® QCS8550 (Proxy) | 7.13 ms | 0 - 195 MB | NPU |
| ConvNext-Base | ONNX | float | Qualcomm® QCS9075 | 11.126 ms | 0 - 4 MB | NPU |
| ConvNext-Base | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 4.134 ms | 0 - 286 MB | NPU |
| ConvNext-Base | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 2.596 ms | 0 - 224 MB | NPU |
| ConvNext-Base | ONNX | w8a16 | Snapdragon® 8 Elite Mobile | 3.199 ms | 0 - 209 MB | NPU |
| ConvNext-Base | ONNX | w8a16 | Snapdragon® X2 Elite | 2.766 ms | 90 - 90 MB | NPU |
| ConvNext-Base | ONNX | w8a16 | Snapdragon® X Elite | 6.459 ms | 90 - 90 MB | NPU |
| ConvNext-Base | ONNX | w8a16 | Snapdragon® X Elite | 6.459 ms | 90 - 90 MB | NPU |
| ConvNext-Base | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 4.375 ms | 0 - 274 MB | NPU |
| ConvNext-Base | ONNX | w8a16 | Qualcomm® QCS6490 | 1101.834 ms | 33 - 65 MB | CPU |
| ConvNext-Base | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 6.191 ms | 0 - 102 MB | NPU |
| ConvNext-Base | ONNX | w8a16 | Qualcomm® QCS9075 | 5.906 ms | 0 - 3 MB | NPU |
| ConvNext-Base | ONNX | w8a16 | Qualcomm® QCM6690 | 629.913 ms | 44 - 56 MB | CPU |
| ConvNext-Base | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 3.199 ms | 0 - 209 MB | NPU |
| ConvNext-Base | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 597.698 ms | 48 - 63 MB | CPU |
| ConvNext-Base | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 597.698 ms | 48 - 63 MB | CPU |
| ConvNext-Base | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.479 ms | 1 - 183 MB | NPU |
| ConvNext-Base | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 4.57 ms | 1 - 183 MB | NPU |
| ConvNext-Base | QNN_DLC | float | Snapdragon® X2 Elite | 4.174 ms | 1 - 1 MB | NPU |
| ConvNext-Base | QNN_DLC | float | Snapdragon® X Elite | 8.371 ms | 1 - 1 MB | NPU |
| ConvNext-Base | QNN_DLC | float | Snapdragon® X Elite | 8.371 ms | 1 - 1 MB | NPU |
| ConvNext-Base | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 5.868 ms | 0 - 306 MB | NPU |
| ConvNext-Base | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 41.898 ms | 1 - 180 MB | NPU |
| ConvNext-Base | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 7.939 ms | 1 - 3 MB | NPU |
| ConvNext-Base | QNN_DLC | float | Qualcomm® QCS9075 | 11.979 ms | 1 - 3 MB | NPU |
| ConvNext-Base | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 20.396 ms | 0 - 295 MB | NPU |
| ConvNext-Base | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 4.57 ms | 1 - 183 MB | NPU |
| ConvNext-Base | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 2.525 ms | 0 - 211 MB | NPU |
| ConvNext-Base | QNN_DLC | w8a16 | Snapdragon® 8 Elite Mobile | 3.277 ms | 0 - 194 MB | NPU |
| ConvNext-Base | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 3.127 ms | 0 - 0 MB | NPU |
| ConvNext-Base | QNN_DLC | w8a16 | Snapdragon® X Elite | 6.256 ms | 0 - 0 MB | NPU |
| ConvNext-Base | QNN_DLC | w8a16 | Snapdragon® X Elite | 6.256 ms | 0 - 0 MB | NPU |
| ConvNext-Base | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 4.075 ms | 0 - 249 MB | NPU |
| ConvNext-Base | QNN_DLC | w8a16 | Qualcomm® QCS6490 | 20.538 ms | 0 - 2 MB | NPU |
| ConvNext-Base | QNN_DLC | w8a16 | Qualcomm® QCS8275 (Proxy) | 14.609 ms | 0 - 202 MB | NPU |
| ConvNext-Base | QNN_DLC | w8a16 | Qualcomm® QCS8550 (Proxy) | 5.873 ms | 0 - 269 MB | NPU |
| ConvNext-Base | QNN_DLC | w8a16 | Qualcomm® QCS9075 | 6.124 ms | 0 - 2 MB | NPU |
| ConvNext-Base | QNN_DLC | w8a16 | Qualcomm® QCM6690 | 68.273 ms | 0 - 401 MB | NPU |
| ConvNext-Base | QNN_DLC | w8a16 | Qualcomm® QCS8450 (Proxy) | 9.158 ms | 0 - 249 MB | NPU |
| ConvNext-Base | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 3.277 ms | 0 - 194 MB | NPU |
| ConvNext-Base | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 7.739 ms | 0 - 252 MB | NPU |
| ConvNext-Base | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 7.739 ms | 0 - 252 MB | NPU |
| ConvNext-Base | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.151 ms | 0 - 179 MB | NPU |
| ConvNext-Base | TFLITE | float | Snapdragon® 8 Elite Mobile | 4.089 ms | 0 - 179 MB | NPU |
| ConvNext-Base | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 5.457 ms | 0 - 298 MB | NPU |
| ConvNext-Base | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 40.988 ms | 0 - 173 MB | NPU |
| ConvNext-Base | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 7.243 ms | 0 - 2 MB | NPU |
| ConvNext-Base | TFLITE | float | Qualcomm® QCS9075 | 11.207 ms | 0 - 177 MB | NPU |
| ConvNext-Base | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 19.708 ms | 0 - 291 MB | NPU |
| ConvNext-Base | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 4.089 ms | 0 - 179 MB | NPU |
License
- The license for the original implementation of ConvNext-Base can be found here.
References
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.