Mobile-VIT

0 downloads 0 likes on ModelScope updated

Model card

Mobile-VIT: Optimized for Qualcomm Devices

MobileVit 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 Mobile-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
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 Mobile-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 Mobile-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): 21.4 MB
  • Model size (w8a16): 6.56 MB
  • Number of parameters: 5.57M

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
Mobile-VIT ONNX float Snapdragon® X2 Elite 2.271 ms 2 - 2 MB NPU
Mobile-VIT ONNX float Snapdragon® X Elite 4.579 ms 12 - 12 MB NPU
Mobile-VIT ONNX float Snapdragon® 8 Gen 3 Mobile 3.025 ms 0 - 93 MB NPU
Mobile-VIT ONNX float Snapdragon® 8 Gen 1 Mobile 6.606 ms 1 - 99 MB NPU
Mobile-VIT ONNX float Qualcomm® Dragonwing™ IQ-8275 4.687 ms 1 - 4 MB NPU
Mobile-VIT ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 4.345 ms 0 - 51 MB NPU
Mobile-VIT ONNX float Qualcomm® QCS8450 6.606 ms 1 - 99 MB NPU
Mobile-VIT ONNX float Qualcomm® Dragonwing™ IQ-9075 4.989 ms 1 - 4 MB NPU
Mobile-VIT ONNX float Qualcomm® Dragonwing™ IQ-X7181 4.579 ms 12 - 12 MB NPU
Mobile-VIT ONNX float Qualcomm® Dragonwing™ Q-8750 2.451 ms 0 - 74 MB NPU
Mobile-VIT ONNX float Snapdragon® 8 Elite Mobile 2.451 ms 0 - 74 MB NPU
Mobile-VIT ONNX float Snapdragon® 8 Elite Gen 5 Mobile 2.057 ms 1 - 66 MB NPU
Mobile-VIT ONNX w8a16 Snapdragon® X2 Elite 1.832 ms 1 - 1 MB NPU
Mobile-VIT ONNX w8a16 Snapdragon® X Elite 4.418 ms 6 - 6 MB NPU
Mobile-VIT ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 2.821 ms 0 - 117 MB NPU
Mobile-VIT ONNX w8a16 Snapdragon® 8 Gen 1 Mobile 4.844 ms 0 - 115 MB NPU
Mobile-VIT ONNX w8a16 Qualcomm® Dragonwing™ QCS6490 13.469 ms 0 - 3 MB NPU
Mobile-VIT ONNX w8a16 Qualcomm® Dragonwing™ IQ-8275 3.869 ms 0 - 4 MB NPU
Mobile-VIT ONNX w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 4.223 ms 0 - 9 MB NPU
Mobile-VIT ONNX w8a16 Qualcomm® QCS8450 4.844 ms 0 - 115 MB NPU
Mobile-VIT ONNX w8a16 Qualcomm® Dragonwing™ IQ-9075 4.483 ms 0 - 3 MB NPU
Mobile-VIT ONNX w8a16 Qualcomm® Dragonwing™ IQ-X7181 4.418 ms 6 - 6 MB NPU
Mobile-VIT ONNX w8a16 Qualcomm® Dragonwing™ Q-6690 19.512 ms 0 - 217 MB NPU
Mobile-VIT ONNX w8a16 Qualcomm® Dragonwing™ Q-7790 4.509 ms 0 - 234 MB NPU
Mobile-VIT ONNX w8a16 Qualcomm® Dragonwing™ Q-8750 2.066 ms 0 - 86 MB NPU
Mobile-VIT ONNX w8a16 Snapdragon® 8 Elite Mobile 2.066 ms 0 - 86 MB NPU
Mobile-VIT ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile 1.753 ms 0 - 87 MB NPU
Mobile-VIT ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 4.509 ms 0 - 234 MB NPU
Mobile-VIT QNN_DLC float Snapdragon® X2 Elite 2.49 ms 1 - 1 MB NPU
Mobile-VIT QNN_DLC float Snapdragon® X Elite 4.608 ms 1 - 1 MB NPU
Mobile-VIT QNN_DLC float Snapdragon® 8 Gen 3 Mobile 2.936 ms 0 - 86 MB NPU
Mobile-VIT QNN_DLC float Snapdragon® 8 Gen 1 Mobile 6.408 ms 0 - 90 MB NPU
Mobile-VIT QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 4.503 ms 1 - 4 MB NPU
Mobile-VIT QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 4.236 ms 1 - 3 MB NPU
Mobile-VIT QNN_DLC float Qualcomm® SA8775P 5.01 ms 1 - 66 MB NPU
Mobile-VIT QNN_DLC float Qualcomm® SA8650P 5.01 ms 1 - 66 MB NPU
Mobile-VIT QNN_DLC float Qualcomm® SA8255P 5.01 ms 1 - 66 MB NPU
Mobile-VIT QNN_DLC float Qualcomm® QCS8450 6.408 ms 0 - 90 MB NPU
Mobile-VIT QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 4.81 ms 3 - 5 MB NPU
Mobile-VIT QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 4.608 ms 1 - 1 MB NPU
Mobile-VIT QNN_DLC float Qualcomm® Dragonwing™ Q-8750 2.385 ms 1 - 56 MB NPU
Mobile-VIT QNN_DLC float Qualcomm® SA7255P 11.192 ms 1 - 57 MB NPU
Mobile-VIT QNN_DLC float Qualcomm® SA8295P 7.362 ms 1 - 61 MB NPU
Mobile-VIT QNN_DLC float Snapdragon® 8 Elite Mobile 2.385 ms 1 - 56 MB NPU
Mobile-VIT QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 2.014 ms 1 - 61 MB NPU
Mobile-VIT QNN_DLC w8a16 Snapdragon® X2 Elite 2.145 ms 0 - 0 MB NPU
Mobile-VIT QNN_DLC w8a16 Snapdragon® X Elite 4.9 ms 0 - 0 MB NPU
Mobile-VIT QNN_DLC w8a16 Snapdragon® 8 Gen 3 Mobile 2.954 ms 0 - 94 MB NPU
Mobile-VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-8275 4.024 ms 0 - 3 MB NPU
Mobile-VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 4.477 ms 0 - 14 MB NPU
Mobile-VIT QNN_DLC w8a16 Qualcomm® SA8775P 4.919 ms 0 - 70 MB NPU
Mobile-VIT QNN_DLC w8a16 Qualcomm® SA8650P 4.919 ms 0 - 70 MB NPU
Mobile-VIT QNN_DLC w8a16 Qualcomm® SA8255P 4.919 ms 0 - 70 MB NPU
Mobile-VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-9075 4.721 ms 0 - 2 MB NPU
Mobile-VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-X7181 4.9 ms 0 - 0 MB NPU
Mobile-VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-6690 17.192 ms 0 - 223 MB NPU
Mobile-VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-7790 4.827 ms 0 - 214 MB NPU
Mobile-VIT QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-8750 2.185 ms 0 - 75 MB NPU
Mobile-VIT QNN_DLC w8a16 Qualcomm® SA7255P 8.031 ms 0 - 70 MB NPU
Mobile-VIT QNN_DLC w8a16 Snapdragon® 8 Elite Mobile 2.185 ms 0 - 75 MB NPU
Mobile-VIT QNN_DLC w8a16 Snapdragon® 8 Elite Gen 5 Mobile 1.811 ms 0 - 74 MB NPU
Mobile-VIT QNN_DLC w8a16 Snapdragon® 7 Gen 4 Mobile 4.827 ms 0 - 214 MB NPU
Mobile-VIT TFLITE float Snapdragon® 8 Gen 3 Mobile 2.949 ms 0 - 102 MB NPU
Mobile-VIT TFLITE float Snapdragon® 8 Gen 1 Mobile 6.452 ms 0 - 104 MB NPU
Mobile-VIT TFLITE float Qualcomm® Dragonwing™ IQ-8275 4.502 ms 0 - 16 MB NPU
Mobile-VIT TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 4.207 ms 0 - 2 MB NPU
Mobile-VIT TFLITE float Qualcomm® SA8775P 5.063 ms 0 - 80 MB NPU
Mobile-VIT TFLITE float Qualcomm® SA8650P 5.063 ms 0 - 80 MB NPU
Mobile-VIT TFLITE float Qualcomm® SA8255P 5.063 ms 0 - 80 MB NPU
Mobile-VIT TFLITE float Qualcomm® QCS8450 6.452 ms 0 - 104 MB NPU
Mobile-VIT TFLITE float Qualcomm® Dragonwing™ IQ-9075 4.79 ms 0 - 15 MB NPU
Mobile-VIT TFLITE float Qualcomm® Dragonwing™ Q-8750 2.386 ms 0 - 77 MB NPU
Mobile-VIT TFLITE float Qualcomm® SA7255P 11.208 ms 0 - 79 MB NPU
Mobile-VIT TFLITE float Qualcomm® SA8295P 7.361 ms 0 - 64 MB NPU
Mobile-VIT TFLITE float Snapdragon® 8 Elite Mobile 2.386 ms 0 - 77 MB NPU
Mobile-VIT TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 2.011 ms 0 - 77 MB NPU

License

  • The license for the original implementation of Mobile-VIT can be found here.

References

Community