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
QNN_DLC w8a8_mixed_int16 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 3.86 ms 2 - 2 MB NPU
VIT ONNX float Snapdragon® X Elite 9.709 ms 170 - 170 MB NPU
VIT ONNX float Snapdragon® 8 Gen 3 Mobile 6.469 ms 0 - 300 MB NPU
VIT ONNX float Snapdragon® 8 Gen 1 Mobile 14.097 ms 0 - 297 MB NPU
VIT ONNX float Qualcomm® Dragonwing™ IQ-8275 12.78 ms 0 - 4 MB NPU
VIT ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 9.245 ms 0 - 193 MB NPU
VIT ONNX float Qualcomm® QCS8450 14.097 ms 0 - 297 MB NPU
VIT ONNX float Qualcomm® Dragonwing™ IQ-9075 11.949 ms 1 - 4 MB NPU
VIT ONNX float Qualcomm® Dragonwing™ IQ-X7181 9.709 ms 170 - 170 MB NPU
VIT ONNX float Qualcomm® Dragonwing™ Q-8750 4.592 ms 0 - 170 MB NPU
VIT ONNX float Snapdragon® 8 Elite Mobile 4.592 ms 0 - 170 MB NPU
VIT ONNX float Snapdragon® 8 Elite Gen 5 Mobile 3.544 ms 0 - 158 MB NPU
VIT ONNX w8a16 Snapdragon® X2 Elite 3.572 ms 1 - 1 MB NPU
VIT ONNX w8a16 Snapdragon® X Elite 8.19 ms 86 - 86 MB NPU
VIT ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 5.343 ms 0 - 320 MB NPU
VIT ONNX w8a16 Snapdragon® 8 Gen 1 Mobile 9.332 ms 0 - 322 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ QCS6490 38.497 ms 0 - 3 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ IQ-8275 7.445 ms 0 - 4 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 7.795 ms 0 - 3 MB NPU
VIT ONNX w8a16 Qualcomm® QCS8450 9.332 ms 0 - 322 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ IQ-9075 7.927 ms 0 - 3 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ IQ-X7181 8.19 ms 86 - 86 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ Q-6690 112.719 ms 0 - 464 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ Q-7790 9.696 ms 0 - 403 MB NPU
VIT ONNX w8a16 Qualcomm® Dragonwing™ Q-8750 4.081 ms 0 - 263 MB NPU
VIT ONNX w8a16 Snapdragon® 8 Elite Mobile 4.081 ms 0 - 263 MB NPU
VIT ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile 3.091 ms 0 - 267 MB NPU
VIT ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 9.696 ms 0 - 403 MB NPU
VIT ONNX w8a8 Snapdragon® X2 Elite 15.354 ms 1 - 1 MB NPU
VIT ONNX w8a8 Snapdragon® X Elite 10.946 ms 86 - 86 MB NPU
VIT ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 7.055 ms 0 - 335 MB NPU
VIT ONNX w8a8 Snapdragon® 8 Gen 1 Mobile 14.568 ms 0 - 342 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ QCS6490 74.031 ms 0 - 3 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ IQ-8275 10.17 ms 0 - 4 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 10.468 ms 0 - 98 MB NPU
VIT ONNX w8a8 Qualcomm® QCS8450 14.568 ms 0 - 342 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ IQ-9075 9.903 ms 0 - 3 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ IQ-X7181 10.946 ms 86 - 86 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ Q-6690 212.002 ms 0 - 505 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ Q-7790 18.888 ms 0 - 338 MB NPU
VIT ONNX w8a8 Qualcomm® Dragonwing™ Q-8750 5.957 ms 0 - 211 MB NPU
VIT ONNX w8a8 Snapdragon® 8 Elite Mobile 5.957 ms 0 - 211 MB NPU
VIT ONNX w8a8 Snapdragon® 8 Elite Gen 5 Mobile 4.297 ms 0 - 220 MB NPU
VIT ONNX w8a8 Snapdragon® 7 Gen 4 Mobile 18.888 ms 0 - 338 MB NPU
VIT ONNX w8a8_mixed_int16 Snapdragon® X2 Elite 10.713 ms 22 - 22 MB NPU
VIT ONNX w8a8_mixed_int16 Snapdragon® X Elite 19.714 ms 83 - 83 MB NPU
VIT ONNX w8a8_mixed_int16 Snapdragon® 8 Gen 3 Mobile 14.923 ms 16 - 518 MB NPU
VIT ONNX w8a8_mixed_int16 Snapdragon® 8 Gen 1 Mobile 31.574 ms 16 - 485 MB NPU
VIT ONNX w8a8_mixed_int16 Qualcomm® Dragonwing™ IQ-8275 22.533 ms 14 - 17 MB NPU
VIT ONNX w8a8_mixed_int16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 20.342 ms 14 - 18 MB NPU
VIT ONNX w8a8_mixed_int16 Qualcomm® QCS8450 31.574 ms 16 - 485 MB NPU
VIT ONNX w8a8_mixed_int16 Qualcomm® Dragonwing™ IQ-9075 22.113 ms 16 - 18 MB NPU
VIT ONNX w8a8_mixed_int16 Qualcomm® Dragonwing™ IQ-X7181 19.714 ms 83 - 83 MB NPU
VIT ONNX w8a8_mixed_int16 Qualcomm® Dragonwing™ Q-8750 11.764 ms 16 - 386 MB NPU
VIT ONNX w8a8_mixed_int16 Snapdragon® 8 Elite Mobile 11.764 ms 16 - 386 MB NPU
VIT ONNX w8a8_mixed_int16 Snapdragon® 8 Elite Gen 5 Mobile 9.625 ms 16 - 399 MB NPU
VIT QNN_DLC float Snapdragon® X2 Elite 6.978 ms 1 - 1 MB NPU
VIT QNN_DLC float Snapdragon® X Elite 12.669 ms 1 - 1 MB NPU
VIT QNN_DLC float Snapdragon® 8 Gen 3 Mobile 8.268 ms 0 - 303 MB NPU
VIT QNN_DLC float Snapdragon® 8 Gen 1 Mobile 15.455 ms 0 - 303 MB NPU
VIT QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 14.939 ms 1 - 4 MB NPU
VIT QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 11.846 ms 1 - 2 MB NPU
VIT QNN_DLC float Qualcomm® SA8775P 14.649 ms 1 - 169 MB NPU
VIT QNN_DLC float Qualcomm® SA8650P 14.649 ms 1 - 169 MB NPU
VIT QNN_DLC float Qualcomm® SA8255P 14.649 ms 1 - 169 MB NPU
VIT QNN_DLC float Qualcomm® QCS8450 15.455 ms 0 - 303 MB NPU
VIT QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 14.892 ms 1 - 3 MB NPU
VIT QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 12.669 ms 1 - 1 MB NPU
VIT QNN_DLC float Qualcomm® Dragonwing™ Q-8750 6.204 ms 1 - 173 MB NPU
VIT QNN_DLC float Qualcomm® SA7255P 42.295 ms 1 - 168 MB NPU
VIT QNN_DLC float Qualcomm® SA8295P 16.197 ms 1 - 164 MB NPU
VIT QNN_DLC float Snapdragon® 8 Elite Mobile 6.204 ms 1 - 173 MB NPU
VIT QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 5.178 ms 1 - 158 MB NPU
VIT QNN_DLC w8a16 Snapdragon® X2 Elite 5.744 ms 0 - 0 MB NPU
VIT QNN_DLC w8a16 Snapdragon® X Elite 13.235 ms 0 - 0 MB NPU
VIT QNN_DLC w8a16 Snapdragon® 8 Gen 3 Mobile 8.698 ms 0 - 338 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-8275 11.779 ms 0 - 3 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 12.392 ms 0 - 2 MB NPU
VIT QNN_DLC w8a16 Qualcomm® SA8775P 12.776 ms 0 - 277 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 12.858 ms 0 - 2 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-X7181 13.235 ms 0 - 0 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-6690 154.402 ms 0 - 477 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-7790 18.368 ms 0 - 429 MB NPU
VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-8750 6.319 ms 0 - 277 MB NPU
VIT QNN_DLC w8a16 Qualcomm® SA7255P 28.685 ms 0 - 276 MB NPU
VIT QNN_DLC w8a16 Snapdragon® 8 Elite Mobile 6.319 ms 0 - 277 MB NPU
VIT QNN_DLC w8a16 Snapdragon® 8 Elite Gen 5 Mobile 4.782 ms 0 - 281 MB NPU
VIT QNN_DLC w8a16 Snapdragon® 7 Gen 4 Mobile 18.368 ms 0 - 429 MB NPU
VIT QNN_DLC w8a8 Snapdragon® X2 Elite 5.366 ms 0 - 0 MB NPU
VIT QNN_DLC w8a8 Snapdragon® X Elite 11.333 ms 0 - 0 MB NPU
VIT QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 7.075 ms 0 - 323 MB NPU
VIT QNN_DLC w8a8 Snapdragon® 8 Gen 1 Mobile 14.532 ms 0 - 333 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS6490 74.196 ms 0 - 2 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-8275 10.161 ms 0 - 3 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 10.585 ms 0 - 2 MB NPU
VIT QNN_DLC w8a8 Qualcomm® SA8775P 10.361 ms 0 - 201 MB NPU
VIT QNN_DLC w8a8 Qualcomm® SA8650P 10.361 ms 0 - 201 MB NPU
VIT QNN_DLC w8a8 Qualcomm® SA8255P 10.361 ms 0 - 201 MB NPU
VIT QNN_DLC w8a8 Qualcomm® QCS8450 14.532 ms 0 - 333 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-9075 10.035 ms 0 - 2 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-X7181 11.333 ms 0 - 0 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-6690 265.268 ms 0 - 512 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-7790 18.618 ms 0 - 332 MB NPU
VIT QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-8750 6.068 ms 0 - 203 MB NPU
VIT QNN_DLC w8a8 Qualcomm® SA7255P 31.183 ms 0 - 202 MB NPU
VIT QNN_DLC w8a8 Qualcomm® SA8295P 17.423 ms 0 - 211 MB NPU
VIT QNN_DLC w8a8 Snapdragon® 8 Elite Mobile 6.068 ms 0 - 203 MB NPU
VIT QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 Mobile 4.334 ms 0 - 210 MB NPU
VIT QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 18.618 ms 0 - 332 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Snapdragon® X2 Elite 5.571 ms 0 - 0 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Snapdragon® X Elite 12.75 ms 1 - 1 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Snapdragon® 8 Gen 3 Mobile 8.139 ms 0 - 478 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Qualcomm® Dragonwing™ IQ-8275 11.452 ms 0 - 3 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 12.059 ms 0 - 258 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Qualcomm® SA8775P 11.521 ms 0 - 349 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 11.292 ms 2 - 4 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Qualcomm® Dragonwing™ IQ-X7181 12.75 ms 1 - 1 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Qualcomm® Dragonwing™ Q-6690 220.717 ms 0 - 538 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Qualcomm® Dragonwing™ Q-7790 19.558 ms 0 - 372 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Qualcomm® Dragonwing™ Q-8750 6.723 ms 0 - 347 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Qualcomm® SA7255P 32.512 ms 0 - 348 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Snapdragon® 8 Elite Mobile 6.723 ms 0 - 347 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Snapdragon® 8 Elite Gen 5 Mobile 4.717 ms 0 - 343 MB NPU
VIT QNN_DLC w8a8_mixed_int16 Snapdragon® 7 Gen 4 Mobile 19.558 ms 0 - 372 MB NPU
VIT TFLITE float Snapdragon® 8 Gen 3 Mobile 6.802 ms 0 - 308 MB NPU
VIT TFLITE float Snapdragon® 8 Gen 1 Mobile 15.279 ms 0 - 298 MB NPU
VIT TFLITE float Qualcomm® Dragonwing™ IQ-8275 13.271 ms 0 - 173 MB NPU
VIT TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 9.325 ms 0 - 2 MB NPU
VIT TFLITE float Qualcomm® SA8775P 12.577 ms 0 - 170 MB NPU
VIT TFLITE float Qualcomm® SA8650P 12.577 ms 0 - 170 MB NPU
VIT TFLITE float Qualcomm® SA8255P 12.577 ms 0 - 170 MB NPU
VIT TFLITE float Qualcomm® QCS8450 15.279 ms 0 - 298 MB NPU
VIT TFLITE float Qualcomm® Dragonwing™ IQ-9075 12.595 ms 0 - 173 MB NPU
VIT TFLITE float Qualcomm® Dragonwing™ Q-8750 4.871 ms 0 - 180 MB NPU
VIT TFLITE float Qualcomm® SA7255P 38.55 ms 0 - 169 MB NPU
VIT TFLITE float Qualcomm® SA8295P 15.705 ms 0 - 164 MB NPU
VIT TFLITE float Snapdragon® 8 Elite Mobile 4.871 ms 0 - 180 MB NPU
VIT TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 3.741 ms 0 - 166 MB NPU
VIT TFLITE w8a8 Snapdragon® 8 Gen 3 Mobile 9.524 ms 0 - 473 MB NPU
VIT TFLITE w8a8 Snapdragon® 8 Gen 1 Mobile 21.784 ms 0 - 464 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ QCS6490 136.865 ms 1 - 100 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ IQ-8275 13.742 ms 0 - 87 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 13.793 ms 0 - 2 MB NPU
VIT TFLITE w8a8 Qualcomm® SA8775P 13.761 ms 0 - 342 MB NPU
VIT TFLITE w8a8 Qualcomm® SA8650P 13.761 ms 0 - 342 MB NPU
VIT TFLITE w8a8 Qualcomm® SA8255P 13.761 ms 0 - 342 MB NPU
VIT TFLITE w8a8 Qualcomm® QCS8450 21.784 ms 0 - 464 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ IQ-9075 13.65 ms 0 - 87 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ Q-6690 202.274 ms 1 - 259 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ Q-7790 30.833 ms 1 - 188 MB NPU
VIT TFLITE w8a8 Qualcomm® Dragonwing™ Q-8750 7.775 ms 0 - 341 MB NPU
VIT TFLITE w8a8 Qualcomm® SA7255P 37.333 ms 0 - 341 MB NPU
VIT TFLITE w8a8 Qualcomm® SA8295P 21.362 ms 0 - 333 MB NPU
VIT TFLITE w8a8 Snapdragon® 8 Elite Mobile 7.775 ms 0 - 341 MB NPU
VIT TFLITE w8a8 Snapdragon® 8 Elite Gen 5 Mobile 5.553 ms 0 - 349 MB NPU
VIT TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 30.833 ms 1 - 188 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