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 2.50, ONNX Runtime 1.27.1 Download
ONNX w8a16 Universal QAIRT 2.50, ONNX Runtime 1.27.1 Download
ONNX w8a8 Universal QAIRT 2.50, ONNX Runtime 1.27.1 Download
ONNX w8a8_mixed_int16 Universal QAIRT 2.50, ONNX Runtime 1.27.1 Download
QNN_DLC float Universal QAIRT 2.50 Download
QNN_DLC w8a16 Universal QAIRT 2.50 Download
QNN_DLC w8a8 Universal QAIRT 2.50 Download
QNN_DLC w8a8_mixed_int16 Universal QAIRT 2.50 Download
TFLITE float Universal QAIRT 2.50 Download
TFLITE w8a8 Universal QAIRT 2.50 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® 8 Elite Gen 5 For Galaxy Mobile 3.457 ms 0 - 160 MB NPU
VIT ONNX float Snapdragon® 8 Elite For Galaxy Mobile 4.576 ms 0 - 159 MB NPU
VIT ONNX float Snapdragon® X2 Elite 3.836 ms 2 - 2 MB NPU
VIT ONNX float Snapdragon® X Elite 9.652 ms 171 - 171 MB NPU
VIT ONNX float Snapdragon® 8 Gen 3 Mobile 6.449 ms 0 - 287 MB NPU
VIT ONNX float Snapdragon® 8 Gen 1 Mobile 13.632 ms 1 - 285 MB NPU
VIT ONNX float Qualcomm® Dragonwing™ IQ-8275 12.761 ms 0 - 5 MB NPU
VIT ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 9.175 ms 0 - 194 MB NPU
VIT ONNX float Qualcomm® QCS8450 13.632 ms 1 - 285 MB NPU
VIT ONNX float Qualcomm® Dragonwing™ IQ-9075 12.108 ms 1 - 4 MB NPU
VIT ONNX float Qualcomm® Dragonwing™ IQ-X7181 9.652 ms 171 - 171 MB NPU
VIT ONNX float Qualcomm® Dragonwing™ Q-8750 4.576 ms 0 - 159 MB NPU
VIT ONNX w8a16 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 2.981 ms 0 - 224 MB NPU
VIT ONNX w8a16 Snapdragon® 8 Elite For Galaxy Mobile 4.124 ms 0 - 222 MB NPU
VIT ONNX w8a16 Snapdragon® X2 Elite 3.467 ms 1 - 1 MB NPU
VIT ONNX w8a16 Snapdragon® X Elite 8.062 ms 86 - 86 MB NPU
VIT ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 5.518 ms 0 - 325 MB NPU
VIT ONNX w8a16 Snapdragon® 8 Gen 1 Mobile 8.687 ms 0 - 327 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ QCS6490 30.519 ms 0 - 4 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ IQ-8275 7.604 ms 0 - 5 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 7.69 ms 0 - 98 MB NPU
VIT ONNX w8a16 Qualcomm® QCS8450 8.687 ms 0 - 327 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ IQ-9075 7.799 ms 0 - 4 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ IQ-X7181 8.062 ms 86 - 86 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ Q-6690 82.973 ms 0 - 438 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ Q-7790 9.645 ms 0 - 394 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ Q-8750 4.124 ms 0 - 222 MB NPU
VIT ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 9.645 ms 0 - 394 MB NPU
VIT ONNX w8a8 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 4.254 ms 0 - 221 MB NPU
VIT ONNX w8a8 Snapdragon® 8 Elite For Galaxy Mobile 5.836 ms 0 - 214 MB NPU
VIT ONNX w8a8 Snapdragon® X2 Elite 4.889 ms 1 - 1 MB NPU
VIT ONNX w8a8 Snapdragon® X Elite 10.877 ms 86 - 86 MB NPU
VIT ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 6.944 ms 0 - 334 MB NPU
VIT ONNX w8a8 Snapdragon® 8 Gen 1 Mobile 13.743 ms 0 - 345 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ QCS6490 57.56 ms 0 - 3 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ IQ-8275 10.119 ms 0 - 4 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 10.394 ms 0 - 99 MB NPU
VIT ONNX w8a8 Qualcomm® QCS8450 13.743 ms 0 - 345 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ IQ-9075 9.84 ms 0 - 4 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ IQ-X7181 10.877 ms 86 - 86 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ Q-6690 215.695 ms 0 - 458 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ Q-7790 19.893 ms 0 - 330 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ Q-8750 5.836 ms 0 - 214 MB NPU
VIT ONNX w8a8 Snapdragon® 7 Gen 4 Mobile 19.893 ms 0 - 330 MB NPU
VIT ONNX w8a8_mixed_int16 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 11.533 ms 40 - 391 MB NPU
VIT ONNX w8a8_mixed_int16 Snapdragon® 8 Elite For Galaxy Mobile 13.807 ms 43 - 379 MB NPU
VIT ONNX w8a8_mixed_int16 Snapdragon® X2 Elite 13.15 ms 67 - 67 MB NPU
VIT ONNX w8a8_mixed_int16 Snapdragon® X Elite 23.304 ms 82 - 82 MB NPU
VIT ONNX w8a8_mixed_int16 Snapdragon® 8 Gen 3 Mobile 17.443 ms 41 - 503 MB NPU
VIT ONNX w8a8_mixed_int16 Snapdragon® 8 Gen 1 Mobile 30.349 ms 42 - 497 MB NPU
VIT ONNX w8a8_mixed_int16 Qualcomm® Dragonwing™ IQ-8275 27.841 ms 38 - 46 MB NPU
VIT ONNX w8a8_mixed_int16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 24.301 ms 38 - 41 MB NPU
VIT ONNX w8a8_mixed_int16 Qualcomm® QCS8450 30.349 ms 42 - 497 MB NPU
VIT ONNX w8a8_mixed_int16 Qualcomm® Dragonwing™ IQ-9075 26.692 ms 42 - 48 MB NPU
VIT ONNX w8a8_mixed_int16 Qualcomm® Dragonwing™ IQ-X7181 23.304 ms 82 - 82 MB NPU
VIT ONNX w8a8_mixed_int16 Qualcomm® Dragonwing™ Q-8750 13.807 ms 43 - 379 MB NPU
VIT QNN_DLC float Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 5.141 ms 0 - 161 MB NPU
VIT QNN_DLC float Snapdragon® 8 Elite For Galaxy Mobile 6.118 ms 0 - 159 MB NPU
VIT QNN_DLC float Snapdragon® X2 Elite 5.674 ms 1 - 1 MB NPU
VIT QNN_DLC float Snapdragon® X Elite 12.622 ms 1 - 1 MB NPU
VIT QNN_DLC float Snapdragon® 8 Gen 3 Mobile 8.204 ms 0 - 289 MB NPU
VIT QNN_DLC float Snapdragon® 8 Gen 1 Mobile 14.858 ms 1 - 288 MB NPU
VIT QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 14.81 ms 1 - 4 MB NPU
VIT QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 11.765 ms 1 - 2 MB NPU
VIT QNN_DLC float Qualcomm® QCS8450 14.858 ms 1 - 288 MB NPU
VIT QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 14.734 ms 1 - 3 MB NPU
VIT QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 12.622 ms 1 - 1 MB NPU
VIT QNN_DLC float Qualcomm® Dragonwing™ Q-8750 6.118 ms 0 - 159 MB NPU
VIT QNN_DLC float Qualcomm® SA7255P 42.14 ms 1 - 157 MB NPU
VIT QNN_DLC float Qualcomm® SA8295P 16.025 ms 1 - 154 MB NPU
VIT QNN_DLC w8a16 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 3.567 ms 0 - 213 MB NPU
VIT QNN_DLC w8a16 Snapdragon® 8 Elite For Galaxy Mobile 4.917 ms 0 - 209 MB NPU
VIT QNN_DLC w8a16 Snapdragon® X2 Elite 4.447 ms 0 - 0 MB NPU
VIT QNN_DLC w8a16 Snapdragon® X Elite 10.075 ms 0 - 0 MB NPU
VIT QNN_DLC w8a16 Snapdragon® 8 Gen 3 Mobile 6.632 ms 0 - 316 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-8275 9.066 ms 0 - 3 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 9.38 ms 0 - 2 MB NPU
VIT QNN_DLC w8a16 Qualcomm® SA8650P 12.776 ms 0 - 277 MB NPU
VIT QNN_DLC w8a16 Qualcomm® SA8255P 12.776 ms 0 - 277 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-9075 9.503 ms 0 - 3 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-X7181 10.075 ms 0 - 0 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-6690 117.44 ms 2 - 432 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-7790 11.808 ms 0 - 380 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-8750 4.917 ms 0 - 209 MB NPU
VIT QNN_DLC w8a16 Qualcomm® SA7255P 22.853 ms 0 - 253 MB NPU
VIT QNN_DLC w8a16 Snapdragon® 7 Gen 4 Mobile 11.808 ms 0 - 380 MB NPU
VIT QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 4.279 ms 0 - 212 MB NPU
VIT QNN_DLC w8a8 Snapdragon® 8 Elite For Galaxy Mobile 5.876 ms 0 - 205 MB NPU
VIT QNN_DLC w8a8 Snapdragon® X2 Elite 5.257 ms 0 - 0 MB NPU
VIT QNN_DLC w8a8 Snapdragon® X Elite 11.227 ms 0 - 0 MB NPU
VIT QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 7.037 ms 0 - 327 MB NPU
VIT QNN_DLC w8a8 Snapdragon® 8 Gen 1 Mobile 13.628 ms 0 - 335 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS6490 57.965 ms 0 - 2 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-8275 10.164 ms 0 - 3 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 10.565 ms 0 - 2 MB NPU
VIT QNN_DLC w8a8 Qualcomm® QCS8450 13.628 ms 0 - 335 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-9075 10.005 ms 0 - 2 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-X7181 11.227 ms 0 - 0 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-6690 229.388 ms 0 - 463 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-7790 19.963 ms 0 - 322 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-8750 5.876 ms 0 - 205 MB NPU
VIT QNN_DLC w8a8 Qualcomm® SA8295P 17.349 ms 0 - 211 MB NPU
VIT QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 19.963 ms 0 - 322 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 4.664 ms 0 - 324 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Snapdragon® 8 Elite For Galaxy Mobile 6.249 ms 0 - 316 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Snapdragon® X2 Elite 5.57 ms 0 - 0 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Snapdragon® X Elite 11.703 ms 0 - 0 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Snapdragon® 8 Gen 3 Mobile 7.433 ms 0 - 438 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Qualcomm® Dragonwing™ IQ-8275 10.638 ms 0 - 3 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 11.044 ms 0 - 2 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Qualcomm® SA8650P 11.521 ms 0 - 349 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Qualcomm® SA8255P 11.521 ms 0 - 349 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Qualcomm® Dragonwing™ IQ-9075 10.499 ms 0 - 2 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Qualcomm® Dragonwing™ IQ-X7181 11.703 ms 0 - 0 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Qualcomm® Dragonwing™ Q-6690 225.416 ms 0 - 490 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Qualcomm® Dragonwing™ Q-7790 21.333 ms 0 - 334 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Qualcomm® Dragonwing™ Q-8750 6.249 ms 0 - 316 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Snapdragon® 7 Gen 4 Mobile 21.333 ms 0 - 334 MB NPU
VIT TFLITE float Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 3.612 ms 0 - 163 MB NPU
VIT TFLITE float Snapdragon® 8 Elite For Galaxy Mobile 4.759 ms 0 - 161 MB NPU
VIT TFLITE float Snapdragon® 8 Gen 3 Mobile 6.745 ms 0 - 290 MB NPU
VIT TFLITE float Snapdragon® 8 Gen 1 Mobile 14.368 ms 0 - 284 MB NPU
VIT TFLITE float Qualcomm® Dragonwing™ IQ-8275 12.998 ms 0 - 174 MB NPU
VIT TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 9.257 ms 0 - 3 MB NPU
VIT TFLITE float Qualcomm® SA8775P 12.461 ms 0 - 158 MB NPU
VIT TFLITE float Qualcomm® SA8650P 12.461 ms 0 - 158 MB NPU
VIT TFLITE float Qualcomm® SA8255P 12.461 ms 0 - 158 MB NPU
VIT TFLITE float Qualcomm® QCS8450 14.368 ms 0 - 284 MB NPU
VIT TFLITE float Qualcomm® Dragonwing™ IQ-9075 12.573 ms 0 - 173 MB NPU
VIT TFLITE float Qualcomm® Dragonwing™ Q-8750 4.759 ms 0 - 161 MB NPU
VIT TFLITE float Qualcomm® SA7255P 38.489 ms 0 - 157 MB NPU
VIT TFLITE float Qualcomm® SA8295P 15.505 ms 0 - 151 MB NPU
VIT TFLITE w8a8 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 5.213 ms 0 - 317 MB NPU
VIT TFLITE w8a8 Snapdragon® 8 Elite For Galaxy Mobile 7.462 ms 0 - 310 MB NPU
VIT TFLITE w8a8 Snapdragon® 8 Gen 3 Mobile 9.229 ms 0 - 444 MB NPU
VIT TFLITE w8a8 Snapdragon® 8 Gen 1 Mobile 18.407 ms 0 - 453 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ QCS6490 148.614 ms 1 - 101 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ IQ-8275 13.1 ms 0 - 88 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 13.12 ms 0 - 2 MB NPU
VIT TFLITE w8a8 Qualcomm® SA8775P 13.123 ms 0 - 311 MB NPU
VIT TFLITE w8a8 Qualcomm® SA8650P 13.123 ms 0 - 311 MB NPU
VIT TFLITE w8a8 Qualcomm® SA8255P 13.123 ms 0 - 311 MB NPU
VIT TFLITE w8a8 Qualcomm® QCS8450 18.407 ms 0 - 453 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ IQ-9075 13.246 ms 0 - 87 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ Q-6690 197.898 ms 2 - 246 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ Q-7790 31.272 ms 2 - 177 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ Q-8750 7.462 ms 0 - 310 MB NPU
VIT TFLITE w8a8 Qualcomm® SA7255P 36.481 ms 0 - 311 MB NPU
VIT TFLITE w8a8 Qualcomm® SA8295P 20.924 ms 0 - 319 MB NPU
VIT TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 31.272 ms 2 - 177 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