ResNeXt50: Optimized for Qualcomm Devices

ResNeXt50 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 ResNeXt50 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 2.45, ONNX Runtime 1.27.1 Download
ONNX w8a8 Universal QAIRT 2.45, ONNX Runtime 1.27.1 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 ResNeXt50 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 ResNeXt50 on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.image_classification

Model Stats:

  • Input resolution: 224x224
  • Model checkpoint: Imagenet
  • Model size (float): 95.4 MB
  • Model size (w8a8): 24.8 MB
  • Number of parameters: 25.0M

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
ResNeXt50 ONNX float Snapdragon® X2 Elite 1.121 ms 2 - 2 MB NPU
ResNeXt50 ONNX float Snapdragon® X Elite 2.216 ms 50 - 50 MB NPU
ResNeXt50 ONNX float Snapdragon® 8 Gen 3 Mobile 1.652 ms 0 - 137 MB NPU
ResNeXt50 ONNX float Snapdragon® 8 Gen 1 Mobile 3.477 ms 0 - 116 MB NPU
ResNeXt50 ONNX float Qualcomm® Dragonwing™ IQ-8275 3.782 ms 0 - 4 MB NPU
ResNeXt50 ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 2.289 ms 0 - 199 MB NPU
ResNeXt50 ONNX float Qualcomm® QCS8450 3.477 ms 0 - 116 MB NPU
ResNeXt50 ONNX float Qualcomm® Dragonwing™ IQ-9075 3.458 ms 0 - 4 MB NPU
ResNeXt50 ONNX float Qualcomm® Dragonwing™ IQ-X7181 2.216 ms 50 - 50 MB NPU
ResNeXt50 ONNX float Qualcomm® Dragonwing™ Q-8750 1.391 ms 0 - 73 MB NPU
ResNeXt50 ONNX float Snapdragon® 8 Elite Mobile 1.391 ms 0 - 73 MB NPU
ResNeXt50 ONNX float Snapdragon® 8 Elite Gen 5 Mobile 1.143 ms 1 - 75 MB NPU
ResNeXt50 ONNX w8a8 Snapdragon® X2 Elite 0.51 ms 1 - 1 MB NPU
ResNeXt50 ONNX w8a8 Snapdragon® X Elite 1.095 ms 26 - 26 MB NPU
ResNeXt50 ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 0.806 ms 0 - 105 MB NPU
ResNeXt50 ONNX w8a8 Snapdragon® 8 Gen 1 Mobile 1.489 ms 0 - 106 MB NPU
ResNeXt50 ONNX w8a8 Qualcomm® Dragonwing™ QCS6490 4.051 ms 0 - 3 MB NPU
ResNeXt50 ONNX w8a8 Qualcomm® Dragonwing™ IQ-8275 1.216 ms 0 - 4 MB NPU
ResNeXt50 ONNX w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.1 ms 0 - 193 MB NPU
ResNeXt50 ONNX w8a8 Qualcomm® QCS8450 1.489 ms 0 - 106 MB NPU
ResNeXt50 ONNX w8a8 Qualcomm® Dragonwing™ IQ-9075 1.33 ms 0 - 3 MB NPU
ResNeXt50 ONNX w8a8 Qualcomm® Dragonwing™ IQ-X7181 1.095 ms 26 - 26 MB NPU
ResNeXt50 ONNX w8a8 Qualcomm® Dragonwing™ Q-6690 7.794 ms 0 - 198 MB NPU
ResNeXt50 ONNX w8a8 Qualcomm® Dragonwing™ Q-7790 1.363 ms 0 - 84 MB NPU
ResNeXt50 ONNX w8a8 Qualcomm® Dragonwing™ Q-8750 0.691 ms 0 - 75 MB NPU
ResNeXt50 ONNX w8a8 Snapdragon® 8 Elite Mobile 0.691 ms 0 - 75 MB NPU
ResNeXt50 ONNX w8a8 Snapdragon® 8 Elite Gen 5 Mobile 0.614 ms 0 - 80 MB NPU
ResNeXt50 ONNX w8a8 Snapdragon® 7 Gen 4 Mobile 1.363 ms 0 - 84 MB NPU
ResNeXt50 QNN_DLC float Snapdragon® X2 Elite 1.434 ms 1 - 1 MB NPU
ResNeXt50 QNN_DLC float Snapdragon® X Elite 2.691 ms 1 - 1 MB NPU
ResNeXt50 QNN_DLC float Snapdragon® 8 Gen 3 Mobile 1.791 ms 0 - 135 MB NPU
ResNeXt50 QNN_DLC float Snapdragon® 8 Gen 1 Mobile 4.019 ms 0 - 115 MB NPU
ResNeXt50 QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 3.83 ms 1 - 4 MB NPU
ResNeXt50 QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 12.025 ms 1 - 72 MB NPU
ResNeXt50 QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 2.54 ms 1 - 3 MB NPU
ResNeXt50 QNN_DLC float Qualcomm® SA8775P 3.836 ms 1 - 73 MB NPU
ResNeXt50 QNN_DLC float Qualcomm® SA8650P 3.836 ms 1 - 73 MB NPU
ResNeXt50 QNN_DLC float Qualcomm® SA8255P 3.836 ms 1 - 73 MB NPU
ResNeXt50 QNN_DLC float Qualcomm® QCS8450 4.019 ms 0 - 115 MB NPU
ResNeXt50 QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 3.589 ms 3 - 5 MB NPU
ResNeXt50 QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 2.691 ms 1 - 1 MB NPU
ResNeXt50 QNN_DLC float Qualcomm® Dragonwing™ Q-8750 1.475 ms 0 - 73 MB NPU
ResNeXt50 QNN_DLC float Qualcomm® SA7255P 12.025 ms 1 - 72 MB NPU
ResNeXt50 QNN_DLC float Qualcomm® SA8295P 4.099 ms 1 - 52 MB NPU
ResNeXt50 QNN_DLC float Snapdragon® 8 Elite Mobile 1.475 ms 0 - 73 MB NPU
ResNeXt50 QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 1.186 ms 1 - 76 MB NPU
ResNeXt50 QNN_DLC w8a8 Snapdragon® X2 Elite 0.64 ms 0 - 0 MB NPU
ResNeXt50 QNN_DLC w8a8 Snapdragon® X Elite 1.242 ms 0 - 0 MB NPU
ResNeXt50 QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 0.779 ms 0 - 101 MB NPU
ResNeXt50 QNN_DLC w8a8 Snapdragon® 8 Gen 1 Mobile 1.468 ms 0 - 101 MB NPU
ResNeXt50 QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS6490 4.169 ms 0 - 2 MB NPU
ResNeXt50 QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-8275 1.157 ms 0 - 3 MB NPU
ResNeXt50 QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-8275 2.43 ms 0 - 74 MB NPU
ResNeXt50 QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.091 ms 0 - 14 MB NPU
ResNeXt50 QNN_DLC w8a8 Qualcomm® SA8775P 1.462 ms 0 - 75 MB NPU
ResNeXt50 QNN_DLC w8a8 Qualcomm® SA8650P 1.462 ms 0 - 75 MB NPU
ResNeXt50 QNN_DLC w8a8 Qualcomm® SA8255P 1.462 ms 0 - 75 MB NPU
ResNeXt50 QNN_DLC w8a8 Qualcomm® QCS8450 1.468 ms 0 - 101 MB NPU
ResNeXt50 QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-9075 1.303 ms 0 - 2 MB NPU
ResNeXt50 QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-X7181 1.242 ms 0 - 0 MB NPU
ResNeXt50 QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-6690 7.764 ms 0 - 196 MB NPU
ResNeXt50 QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-7790 1.375 ms 0 - 77 MB NPU
ResNeXt50 QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-8750 0.645 ms 0 - 76 MB NPU
ResNeXt50 QNN_DLC w8a8 Qualcomm® SA7255P 2.43 ms 0 - 74 MB NPU
ResNeXt50 QNN_DLC w8a8 Qualcomm® SA8295P 1.778 ms 0 - 71 MB NPU
ResNeXt50 QNN_DLC w8a8 Snapdragon® 8 Elite Mobile 0.645 ms 0 - 76 MB NPU
ResNeXt50 QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 Mobile 0.537 ms 0 - 77 MB NPU
ResNeXt50 QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 1.375 ms 0 - 77 MB NPU
ResNeXt50 TFLITE float Snapdragon® 8 Gen 3 Mobile 1.778 ms 0 - 175 MB NPU
ResNeXt50 TFLITE float Snapdragon® 8 Gen 1 Mobile 4.01 ms 0 - 152 MB NPU
ResNeXt50 TFLITE float Qualcomm® Dragonwing™ IQ-8275 3.83 ms 0 - 53 MB NPU
ResNeXt50 TFLITE float Qualcomm® Dragonwing™ IQ-8275 12.039 ms 0 - 111 MB NPU
ResNeXt50 TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 2.499 ms 0 - 2 MB NPU
ResNeXt50 TFLITE float Qualcomm® SA8775P 3.826 ms 0 - 114 MB NPU
ResNeXt50 TFLITE float Qualcomm® SA8650P 3.826 ms 0 - 114 MB NPU
ResNeXt50 TFLITE float Qualcomm® SA8255P 3.826 ms 0 - 114 MB NPU
ResNeXt50 TFLITE float Qualcomm® QCS8450 4.01 ms 0 - 152 MB NPU
ResNeXt50 TFLITE float Qualcomm® Dragonwing™ IQ-9075 3.592 ms 0 - 52 MB NPU
ResNeXt50 TFLITE float Qualcomm® Dragonwing™ Q-8750 1.435 ms 0 - 115 MB NPU
ResNeXt50 TFLITE float Qualcomm® SA7255P 12.039 ms 0 - 111 MB NPU
ResNeXt50 TFLITE float Qualcomm® SA8295P 4.143 ms 0 - 87 MB NPU
ResNeXt50 TFLITE float Snapdragon® 8 Elite Mobile 1.435 ms 0 - 115 MB NPU
ResNeXt50 TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 1.194 ms 0 - 116 MB NPU
ResNeXt50 TFLITE w8a8 Snapdragon® 8 Gen 3 Mobile 0.685 ms 0 - 101 MB NPU
ResNeXt50 TFLITE w8a8 Snapdragon® 8 Gen 1 Mobile 1.281 ms 0 - 97 MB NPU
ResNeXt50 TFLITE w8a8 Qualcomm® Dragonwing™ QCS6490 3.889 ms 0 - 27 MB NPU
ResNeXt50 TFLITE w8a8 Qualcomm® Dragonwing™ IQ-8275 1.015 ms 0 - 28 MB NPU
ResNeXt50 TFLITE w8a8 Qualcomm® Dragonwing™ IQ-8275 2.213 ms 0 - 69 MB NPU
ResNeXt50 TFLITE w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.902 ms 0 - 2 MB NPU
ResNeXt50 TFLITE w8a8 Qualcomm® SA8775P 1.268 ms 0 - 71 MB NPU
ResNeXt50 TFLITE w8a8 Qualcomm® SA8650P 1.268 ms 0 - 71 MB NPU
ResNeXt50 TFLITE w8a8 Qualcomm® SA8255P 1.268 ms 0 - 71 MB NPU
ResNeXt50 TFLITE w8a8 Qualcomm® QCS8450 1.281 ms 0 - 97 MB NPU
ResNeXt50 TFLITE w8a8 Qualcomm® Dragonwing™ IQ-9075 1.089 ms 0 - 27 MB NPU
ResNeXt50 TFLITE w8a8 Qualcomm® Dragonwing™ Q-6690 7.136 ms 0 - 198 MB NPU
ResNeXt50 TFLITE w8a8 Qualcomm® Dragonwing™ Q-7790 1.217 ms 0 - 70 MB NPU
ResNeXt50 TFLITE w8a8 Qualcomm® Dragonwing™ Q-8750 0.55 ms 0 - 63 MB NPU
ResNeXt50 TFLITE w8a8 Qualcomm® SA7255P 2.213 ms 0 - 69 MB NPU
ResNeXt50 TFLITE w8a8 Qualcomm® SA8295P 1.56 ms 0 - 65 MB NPU
ResNeXt50 TFLITE w8a8 Snapdragon® 8 Elite Mobile 0.55 ms 0 - 63 MB NPU
ResNeXt50 TFLITE w8a8 Snapdragon® 8 Elite Gen 5 Mobile 0.493 ms 0 - 69 MB NPU
ResNeXt50 TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 1.217 ms 0 - 70 MB NPU

License

  • The license for the original implementation of ResNeXt50 can be found here.

References

Community

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Paper for qualcomm/ResNeXt50