Image Classification
PyTorch
backbone
android

VIT: Optimized for Qualcomm Devices

VIT 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 VIT 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 w8a16 Universal QAIRT 2.45, ONNX Runtime 1.27.1 Download
ONNX w8a8 Universal QAIRT 2.45, ONNX Runtime 1.27.1 Download
ONNX w8a8_mixed_int16 Universal QAIRT 2.45, ONNX Runtime 1.27.1 Download
QNN_DLC float Universal QAIRT 2.45 Download
QNN_DLC w8a16 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 VIT 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 VIT 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): 330 MB
  • Model size (w8a16): 86.2 MB
  • Model size (w8a8): 83.2 MB
  • Number of parameters: 86.6M

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
VIT ONNX float Snapdragon® X2 Elite 3.869 ms 2 - 2 MB NPU
VIT ONNX float Snapdragon® X Elite 9.697 ms 171 - 171 MB NPU
VIT ONNX float Snapdragon® 8 Gen 3 Mobile 6.479 ms 0 - 302 MB NPU
VIT ONNX float Snapdragon® 8 Gen 1 Mobile 14.151 ms 1 - 299 MB NPU
VIT ONNX float Qualcomm® Dragonwing™ IQ-8275 12.647 ms 0 - 4 MB NPU
VIT ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 9.244 ms 0 - 422 MB NPU
VIT ONNX float Qualcomm® QCS8450 14.151 ms 1 - 299 MB NPU
VIT ONNX float Qualcomm® Dragonwing™ IQ-9075 12.098 ms 0 - 4 MB NPU
VIT ONNX float Qualcomm® Dragonwing™ IQ-X7181 9.697 ms 171 - 171 MB NPU
VIT ONNX float Qualcomm® Dragonwing™ Q-8750 4.636 ms 0 - 170 MB NPU
VIT ONNX float Snapdragon® 8 Elite Mobile 4.636 ms 0 - 170 MB NPU
VIT ONNX float Snapdragon® 8 Elite Gen 5 Mobile 3.564 ms 0 - 159 MB NPU
VIT ONNX w8a16 Snapdragon® X2 Elite 3.572 ms 1 - 1 MB NPU
VIT ONNX w8a16 Snapdragon® X Elite 8.223 ms 85 - 85 MB NPU
VIT ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 5.344 ms 0 - 320 MB NPU
VIT ONNX w8a16 Snapdragon® 8 Gen 1 Mobile 9.255 ms 0 - 323 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ QCS6490 38.332 ms 0 - 3 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ IQ-8275 7.47 ms 0 - 4 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 7.825 ms 0 - 98 MB NPU
VIT ONNX w8a16 Qualcomm® QCS8450 9.255 ms 0 - 323 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ IQ-9075 7.935 ms 0 - 3 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ IQ-X7181 8.223 ms 85 - 85 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ Q-6690 106.969 ms 0 - 465 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ Q-7790 9.753 ms 0 - 401 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ Q-8750 4.091 ms 0 - 264 MB NPU
VIT ONNX w8a16 Snapdragon® 8 Elite Mobile 4.091 ms 0 - 264 MB NPU
VIT ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile 3.089 ms 0 - 266 MB NPU
VIT ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 9.753 ms 0 - 401 MB NPU
VIT ONNX w8a8 Snapdragon® X2 Elite 4.944 ms 1 - 1 MB NPU
VIT ONNX w8a8 Snapdragon® X Elite 10.979 ms 86 - 86 MB NPU
VIT ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 7.058 ms 0 - 334 MB NPU
VIT ONNX w8a8 Snapdragon® 8 Gen 1 Mobile 14.661 ms 0 - 342 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ QCS6490 74.034 ms 0 - 3 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ IQ-8275 10.181 ms 0 - 4 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 10.507 ms 0 - 98 MB NPU
VIT ONNX w8a8 Qualcomm® QCS8450 14.661 ms 0 - 342 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ IQ-9075 9.9 ms 0 - 3 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ IQ-X7181 10.979 ms 86 - 86 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ Q-6690 166.993 ms 0 - 509 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ Q-7790 18.871 ms 0 - 336 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ Q-8750 5.951 ms 0 - 214 MB NPU
VIT ONNX w8a8 Snapdragon® 8 Elite Mobile 5.951 ms 0 - 214 MB NPU
VIT ONNX w8a8 Snapdragon® 8 Elite Gen 5 Mobile 4.301 ms 0 - 221 MB NPU
VIT ONNX w8a8 Snapdragon® 7 Gen 4 Mobile 18.871 ms 0 - 336 MB NPU
VIT ONNX w8a8_mixed_int16 Snapdragon® 8 Gen 3 Mobile 15.976 ms 17 - 499 MB NPU
VIT ONNX w8a8_mixed_int16 Snapdragon® 8 Gen 1 Mobile 31.408 ms 13 - 481 MB NPU
VIT ONNX w8a8_mixed_int16 Qualcomm® Dragonwing™ IQ-8275 24.238 ms 16 - 19 MB NPU
VIT ONNX w8a8_mixed_int16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 22.09 ms 16 - 19 MB NPU
VIT ONNX w8a8_mixed_int16 Qualcomm® QCS8450 31.408 ms 13 - 481 MB NPU
VIT ONNX w8a8_mixed_int16 Qualcomm® Dragonwing™ IQ-9075 26.202 ms 20 - 22 MB NPU
VIT ONNX w8a8_mixed_int16 Qualcomm® Dragonwing™ Q-8750 12.745 ms 93 - 446 MB NPU
VIT ONNX w8a8_mixed_int16 Snapdragon® 8 Elite Mobile 12.745 ms 93 - 446 MB NPU
VIT ONNX w8a8_mixed_int16 Snapdragon® 8 Elite Gen 5 Mobile 10.123 ms 18 - 386 MB NPU
VIT QNN_DLC float Snapdragon® X2 Elite 5.73 ms 1 - 1 MB NPU
VIT QNN_DLC float Snapdragon® X Elite 12.692 ms 1 - 1 MB NPU
VIT QNN_DLC float Snapdragon® 8 Gen 3 Mobile 8.262 ms 0 - 303 MB NPU
VIT QNN_DLC float Snapdragon® 8 Gen 1 Mobile 15.564 ms 0 - 303 MB NPU
VIT QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 14.924 ms 1 - 4 MB NPU
VIT QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 42.297 ms 1 - 168 MB NPU
VIT QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 11.866 ms 1 - 2 MB NPU
VIT QNN_DLC float Qualcomm® SA8775P 14.651 ms 1 - 169 MB NPU
VIT QNN_DLC float Qualcomm® SA8650P 14.651 ms 1 - 169 MB NPU
VIT QNN_DLC float Qualcomm® SA8255P 14.651 ms 1 - 169 MB NPU
VIT QNN_DLC float Qualcomm® QCS8450 15.564 ms 0 - 303 MB NPU
VIT QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 14.839 ms 1 - 3 MB NPU
VIT QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 12.692 ms 1 - 1 MB NPU
VIT QNN_DLC float Qualcomm® Dragonwing™ Q-8750 6.187 ms 1 - 172 MB NPU
VIT QNN_DLC float Qualcomm® SA7255P 42.297 ms 1 - 168 MB NPU
VIT QNN_DLC float Qualcomm® SA8295P 16.192 ms 1 - 164 MB NPU
VIT QNN_DLC float Snapdragon® 8 Elite Mobile 6.187 ms 1 - 172 MB NPU
VIT QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 5.183 ms 1 - 159 MB NPU
VIT QNN_DLC w8a16 Snapdragon® X2 Elite 9.168 ms 0 - 0 MB NPU
VIT QNN_DLC w8a16 Snapdragon® X Elite 13.315 ms 0 - 0 MB NPU
VIT QNN_DLC w8a16 Snapdragon® 8 Gen 3 Mobile 8.688 ms 0 - 336 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-8275 11.821 ms 0 - 3 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-8275 28.652 ms 0 - 276 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 12.404 ms 0 - 2 MB NPU
VIT QNN_DLC w8a16 Qualcomm® SA8775P 12.798 ms 0 - 278 MB NPU
VIT QNN_DLC w8a16 Qualcomm® SA8650P 12.798 ms 0 - 278 MB NPU
VIT QNN_DLC w8a16 Qualcomm® SA8255P 12.798 ms 0 - 278 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-9075 12.627 ms 2 - 4 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-X7181 13.315 ms 0 - 0 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-6690 122.065 ms 0 - 477 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-7790 18.092 ms 0 - 430 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-8750 6.315 ms 0 - 277 MB NPU
VIT QNN_DLC w8a16 Qualcomm® SA7255P 28.652 ms 0 - 276 MB NPU
VIT QNN_DLC w8a16 Snapdragon® 8 Elite Mobile 6.315 ms 0 - 277 MB NPU
VIT QNN_DLC w8a16 Snapdragon® 8 Elite Gen 5 Mobile 4.778 ms 0 - 280 MB NPU
VIT QNN_DLC w8a16 Snapdragon® 7 Gen 4 Mobile 18.092 ms 0 - 430 MB NPU
VIT QNN_DLC w8a8 Snapdragon® X2 Elite 5.371 ms 0 - 0 MB NPU
VIT QNN_DLC w8a8 Snapdragon® X Elite 11.302 ms 0 - 0 MB NPU
VIT QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 7.035 ms 0 - 321 MB NPU
VIT QNN_DLC w8a8 Snapdragon® 8 Gen 1 Mobile 15.37 ms 0 - 333 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS6490 74.159 ms 0 - 2 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-8275 10.163 ms 0 - 3 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-8275 31.163 ms 0 - 202 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 10.613 ms 0 - 2 MB NPU
VIT QNN_DLC w8a8 Qualcomm® SA8775P 10.352 ms 0 - 201 MB NPU
VIT QNN_DLC w8a8 Qualcomm® SA8650P 10.352 ms 0 - 201 MB NPU
VIT QNN_DLC w8a8 Qualcomm® SA8255P 10.352 ms 0 - 201 MB NPU
VIT QNN_DLC w8a8 Qualcomm® QCS8450 15.37 ms 0 - 333 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-9075 10.043 ms 0 - 2 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-X7181 11.302 ms 0 - 0 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-6690 259.059 ms 0 - 508 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-7790 18.558 ms 0 - 331 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-8750 6.012 ms 0 - 203 MB NPU
VIT QNN_DLC w8a8 Qualcomm® SA7255P 31.163 ms 0 - 202 MB NPU
VIT QNN_DLC w8a8 Qualcomm® SA8295P 17.412 ms 0 - 211 MB NPU
VIT QNN_DLC w8a8 Snapdragon® 8 Elite Mobile 6.012 ms 0 - 203 MB NPU
VIT QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 Mobile 4.342 ms 0 - 209 MB NPU
VIT QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 18.558 ms 0 - 331 MB NPU
VIT TFLITE float Snapdragon® 8 Gen 3 Mobile 6.802 ms 0 - 296 MB NPU
VIT TFLITE float Snapdragon® 8 Gen 1 Mobile 15.181 ms 0 - 300 MB NPU
VIT TFLITE float Qualcomm® Dragonwing™ IQ-8275 13.011 ms 0 - 173 MB NPU
VIT TFLITE float Qualcomm® Dragonwing™ IQ-8275 38.525 ms 0 - 169 MB NPU
VIT TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 9.335 ms 0 - 2 MB NPU
VIT TFLITE float Qualcomm® SA8775P 12.566 ms 0 - 169 MB NPU
VIT TFLITE float Qualcomm® SA8650P 12.566 ms 0 - 169 MB NPU
VIT TFLITE float Qualcomm® SA8255P 12.566 ms 0 - 169 MB NPU
VIT TFLITE float Qualcomm® QCS8450 15.181 ms 0 - 300 MB NPU
VIT TFLITE float Qualcomm® Dragonwing™ IQ-9075 12.543 ms 0 - 173 MB NPU
VIT TFLITE float Qualcomm® Dragonwing™ Q-8750 4.84 ms 0 - 178 MB NPU
VIT TFLITE float Qualcomm® SA7255P 38.525 ms 0 - 169 MB NPU
VIT TFLITE float Qualcomm® SA8295P 15.709 ms 0 - 163 MB NPU
VIT TFLITE float Snapdragon® 8 Elite Mobile 4.84 ms 0 - 178 MB NPU
VIT TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 3.742 ms 0 - 166 MB NPU
VIT TFLITE w8a8 Snapdragon® 8 Gen 3 Mobile 9.526 ms 0 - 477 MB NPU
VIT TFLITE w8a8 Snapdragon® 8 Gen 1 Mobile 21.586 ms 0 - 467 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ QCS6490 140.58 ms 1 - 100 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ IQ-8275 13.606 ms 0 - 87 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ IQ-8275 37.357 ms 0 - 341 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 13.765 ms 0 - 2 MB NPU
VIT TFLITE w8a8 Qualcomm® SA8775P 13.786 ms 0 - 342 MB NPU
VIT TFLITE w8a8 Qualcomm® SA8650P 13.786 ms 0 - 342 MB NPU
VIT TFLITE w8a8 Qualcomm® SA8255P 13.786 ms 0 - 342 MB NPU
VIT TFLITE w8a8 Qualcomm® QCS8450 21.586 ms 0 - 467 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ IQ-9075 13.915 ms 0 - 87 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ Q-6690 201.664 ms 2 - 259 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ Q-7790 30.919 ms 1 - 187 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ Q-8750 7.781 ms 0 - 333 MB NPU
VIT TFLITE w8a8 Qualcomm® SA7255P 37.357 ms 0 - 341 MB NPU
VIT TFLITE w8a8 Qualcomm® SA8295P 21.436 ms 0 - 333 MB NPU
VIT TFLITE w8a8 Snapdragon® 8 Elite Mobile 7.781 ms 0 - 333 MB NPU
VIT TFLITE w8a8 Snapdragon® 8 Elite Gen 5 Mobile 5.511 ms 0 - 351 MB NPU
VIT TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 30.919 ms 1 - 187 MB NPU

License

  • The license for the original implementation of VIT 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/VIT