Hugging Face's logo

qualcomm
/
VIT

Image Classification
PyTorch
LiteRT
backbone
android
Model card Files
xet
Community

VIT: Optimized for Mobile Deployment

Imagenet classifier and general purpose backbone

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 model is an implementation of VIT found here.

This repository provides scripts to run VIT on Qualcomm® devices. More details on model performance across various devices, can be found here.

Model Details

  • Model Type: Model_use_case.image_classification
  • Model Stats:
    • Model checkpoint: Imagenet
    • Input resolution: 224x224
    • Number of parameters: 86.6M
    • Model size (float): 330 MB
    • Model size (w8a16): 86.2 MB
    • Model size (w8a8): 83.2 MB
Model Precision Device Chipset Target Runtime Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit Target Model
VIT float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 37.327 ms 0 - 304 MB NPU VIT.tflite
VIT float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN_DLC 40.725 ms 0 - 302 MB NPU VIT.dlc
VIT float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 13.034 ms 0 - 317 MB NPU VIT.tflite
VIT float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN_DLC 19.296 ms 1 - 303 MB NPU VIT.dlc
VIT float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 8.93 ms 0 - 14 MB NPU VIT.tflite
VIT float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN_DLC 11.473 ms 0 - 24 MB NPU VIT.dlc
VIT float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) ONNX 13.27 ms 1 - 28 MB NPU VIT.onnx.zip
VIT float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 11.819 ms 0 - 304 MB NPU VIT.tflite
VIT float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN_DLC 14.398 ms 1 - 302 MB NPU VIT.dlc
VIT float SA7255P ADP Qualcomm® SA7255P TFLITE 37.327 ms 0 - 304 MB NPU VIT.tflite
VIT float SA7255P ADP Qualcomm® SA7255P QNN_DLC 40.725 ms 0 - 302 MB NPU VIT.dlc
VIT float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 8.592 ms 0 - 19 MB NPU VIT.tflite
VIT float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN_DLC 11.459 ms 0 - 21 MB NPU VIT.dlc
VIT float SA8295P ADP Qualcomm® SA8295P TFLITE 14.687 ms 0 - 306 MB NPU VIT.tflite
VIT float SA8295P ADP Qualcomm® SA8295P QNN_DLC 17.28 ms 1 - 306 MB NPU VIT.dlc
VIT float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 8.556 ms 0 - 27 MB NPU VIT.tflite
VIT float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN_DLC 11.473 ms 0 - 22 MB NPU VIT.dlc
VIT float SA8775P ADP Qualcomm® SA8775P TFLITE 11.819 ms 0 - 304 MB NPU VIT.tflite
VIT float SA8775P ADP Qualcomm® SA8775P QNN_DLC 14.398 ms 1 - 302 MB NPU VIT.dlc
VIT float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 6.272 ms 0 - 310 MB NPU VIT.tflite
VIT float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN_DLC 8.04 ms 1 - 313 MB NPU VIT.dlc
VIT float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 9.08 ms 0 - 331 MB NPU VIT.onnx.zip
VIT float Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile TFLITE 4.443 ms 0 - 308 MB NPU VIT.tflite
VIT float Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile QNN_DLC 5.671 ms 1 - 313 MB NPU VIT.dlc
VIT float Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile ONNX 6.364 ms 1 - 325 MB NPU VIT.onnx.zip
VIT float Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile TFLITE 3.795 ms 0 - 308 MB NPU VIT.tflite
VIT float Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile QNN_DLC 4.348 ms 1 - 304 MB NPU VIT.dlc
VIT float Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile ONNX 4.856 ms 1 - 317 MB NPU VIT.onnx.zip
VIT float Snapdragon X Elite CRD Snapdragon® X Elite QNN_DLC 12.189 ms 1126 - 1126 MB NPU VIT.dlc
VIT float Snapdragon X Elite CRD Snapdragon® X Elite ONNX 13.866 ms 171 - 171 MB NPU VIT.onnx.zip
VIT w8a16 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN_DLC 69.629 ms 0 - 195 MB NPU VIT.dlc
VIT w8a16 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN_DLC 26.732 ms 3 - 47 MB NPU VIT.dlc
VIT w8a16 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) ONNX 314.345 ms 32 - 121 MB NPU VIT.onnx.zip
VIT w8a16 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN_DLC 24.548 ms 0 - 188 MB NPU VIT.dlc
VIT w8a16 RB3 Gen 2 (Proxy) Qualcomm® QCS6490 (Proxy) ONNX 524.615 ms 82 - 100 MB CPU VIT.onnx.zip
VIT w8a16 RB5 (Proxy) Qualcomm® QCS8250 (Proxy) ONNX 548.251 ms 84 - 100 MB CPU VIT.onnx.zip
VIT w8a16 SA7255P ADP Qualcomm® SA7255P QNN_DLC 69.629 ms 0 - 195 MB NPU VIT.dlc
VIT w8a16 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN_DLC 26.687 ms 0 - 44 MB NPU VIT.dlc
VIT w8a16 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN_DLC 26.727 ms 0 - 46 MB NPU VIT.dlc
VIT w8a16 SA8775P ADP Qualcomm® SA8775P QNN_DLC 24.548 ms 0 - 188 MB NPU VIT.dlc
VIT w8a16 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN_DLC 20.318 ms 0 - 200 MB NPU VIT.dlc
VIT w8a16 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 266.191 ms 55 - 92 MB NPU VIT.onnx.zip
VIT w8a16 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile QNN_DLC 15.988 ms 0 - 190 MB NPU VIT.dlc
VIT w8a16 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile ONNX 230.044 ms 55 - 95 MB NPU VIT.onnx.zip
VIT w8a16 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile QNN_DLC 12.942 ms 0 - 194 MB NPU VIT.dlc
VIT w8a16 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile ONNX 204.577 ms 53 - 96 MB NPU VIT.onnx.zip
VIT w8a16 Snapdragon X Elite CRD Snapdragon® X Elite QNN_DLC 26.526 ms 309 - 309 MB NPU VIT.dlc
VIT w8a16 Snapdragon X Elite CRD Snapdragon® X Elite ONNX 169.231 ms 62 - 62 MB NPU VIT.onnx.zip
VIT w8a8 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 15.527 ms 0 - 46 MB NPU VIT.tflite
VIT w8a8 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN_DLC 29.908 ms 0 - 150 MB NPU VIT.dlc
VIT w8a8 QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 8.157 ms 0 - 54 MB NPU VIT.tflite
VIT w8a8 QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN_DLC 13.244 ms 0 - 268 MB NPU VIT.dlc
VIT w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 7.441 ms 0 - 22 MB NPU VIT.tflite
VIT w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN_DLC 10.195 ms 0 - 23 MB NPU VIT.dlc
VIT w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) ONNX 288.901 ms 32 - 121 MB NPU VIT.onnx.zip
VIT w8a8 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 35.33 ms 0 - 46 MB NPU VIT.tflite
VIT w8a8 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN_DLC 9.538 ms 0 - 150 MB NPU VIT.dlc
VIT w8a8 RB3 Gen 2 (Proxy) Qualcomm® QCS6490 (Proxy) TFLITE 89.098 ms 2 - 45 MB NPU VIT.tflite
VIT w8a8 RB3 Gen 2 (Proxy) Qualcomm® QCS6490 (Proxy) QNN_DLC 77.03 ms 0 - 656 MB NPU VIT.dlc
VIT w8a8 RB3 Gen 2 (Proxy) Qualcomm® QCS6490 (Proxy) ONNX 434.397 ms 66 - 83 MB CPU VIT.onnx.zip
VIT w8a8 RB5 (Proxy) Qualcomm® QCS8250 (Proxy) ONNX 482.665 ms 72 - 84 MB CPU VIT.onnx.zip
VIT w8a8 SA7255P ADP Qualcomm® SA7255P TFLITE 15.527 ms 0 - 46 MB NPU VIT.tflite
VIT w8a8 SA7255P ADP Qualcomm® SA7255P QNN_DLC 29.908 ms 0 - 150 MB NPU VIT.dlc
VIT w8a8 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 7.515 ms 0 - 105 MB NPU VIT.tflite
VIT w8a8 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN_DLC 10.218 ms 0 - 22 MB NPU VIT.dlc
VIT w8a8 SA8295P ADP Qualcomm® SA8295P TFLITE 9.774 ms 0 - 48 MB NPU VIT.tflite
VIT w8a8 SA8295P ADP Qualcomm® SA8295P QNN_DLC 16.411 ms 0 - 164 MB NPU VIT.dlc
VIT w8a8 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 7.444 ms 0 - 25 MB NPU VIT.tflite
VIT w8a8 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN_DLC 10.233 ms 0 - 23 MB NPU VIT.dlc
VIT w8a8 SA8775P ADP Qualcomm® SA8775P TFLITE 35.33 ms 0 - 46 MB NPU VIT.tflite
VIT w8a8 SA8775P ADP Qualcomm® SA8775P QNN_DLC 9.538 ms 0 - 150 MB NPU VIT.dlc
VIT w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 5.252 ms 0 - 52 MB NPU VIT.tflite
VIT w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN_DLC 7.003 ms 0 - 156 MB NPU VIT.dlc
VIT w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 253.388 ms 56 - 94 MB NPU VIT.onnx.zip
VIT w8a8 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile TFLITE 4.102 ms 0 - 53 MB NPU VIT.tflite
VIT w8a8 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile QNN_DLC 5.886 ms 0 - 151 MB NPU VIT.dlc
VIT w8a8 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile ONNX 207.142 ms 56 - 95 MB NPU VIT.onnx.zip
VIT w8a8 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile TFLITE 3.471 ms 0 - 54 MB NPU VIT.tflite
VIT w8a8 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile QNN_DLC 3.782 ms 0 - 257 MB NPU VIT.dlc
VIT w8a8 Snapdragon X Elite CRD Snapdragon® X Elite QNN_DLC 10.763 ms 433 - 433 MB NPU VIT.dlc
VIT w8a8 Snapdragon X Elite CRD Snapdragon® X Elite ONNX 183.277 ms 60 - 60 MB NPU VIT.onnx.zip
VIT w8a8_mixed_int16 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) ONNX 355.411 ms 43 - 147 MB NPU VIT.onnx.zip
VIT w8a8_mixed_int16 RB3 Gen 2 (Proxy) Qualcomm® QCS6490 (Proxy) ONNX 456.562 ms 86 - 105 MB CPU VIT.onnx.zip
VIT w8a8_mixed_int16 RB5 (Proxy) Qualcomm® QCS8250 (Proxy) ONNX 460.802 ms 86 - 104 MB CPU VIT.onnx.zip
VIT w8a8_mixed_int16 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 294.86 ms 79 - 122 MB NPU VIT.onnx.zip
VIT w8a8_mixed_int16 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile ONNX 254.105 ms 78 - 120 MB NPU VIT.onnx.zip
VIT w8a8_mixed_int16 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile ONNX 240.281 ms 79 - 122 MB NPU VIT.onnx.zip
VIT w8a8_mixed_int16 Snapdragon X Elite CRD Snapdragon® X Elite ONNX 347.308 ms 130 - 130 MB NPU VIT.onnx.zip

Installation

Install the package via pip:

pip install qai-hub-models

Configure Qualcomm® AI Hub to run this model on a cloud-hosted device

Sign-in to Qualcomm® AI Hub with your Qualcomm® ID. Once signed in navigate to Account -> Settings -> API Token.

With this API token, you can configure your client to run models on the cloud hosted devices.

qai-hub configure --api_token API_TOKEN

Navigate to docs for more information.

Demo off target

The package contains a simple end-to-end demo that downloads pre-trained weights and runs this model on a sample input.

python -m qai_hub_models.models.vit.demo

The above demo runs a reference implementation of pre-processing, model inference, and post processing.

NOTE: If you want running in a Jupyter Notebook or Google Colab like environment, please add the following to your cell (instead of the above).

%run -m qai_hub_models.models.vit.demo

Run model on a cloud-hosted device

In addition to the demo, you can also run the model on a cloud-hosted Qualcomm® device. This script does the following:

  • Performance check on-device on a cloud-hosted device
  • Downloads compiled assets that can be deployed on-device for Android.
  • Accuracy check between PyTorch and on-device outputs.
python -m qai_hub_models.models.vit.export

How does this work?

This export script leverages Qualcomm® AI Hub to optimize, validate, and deploy this model on-device. Lets go through each step below in detail:

Step 1: Compile model for on-device deployment

To compile a PyTorch model for on-device deployment, we first trace the model in memory using the jit.trace and then call the submit_compile_job API.

import torch

import qai_hub as hub
from qai_hub_models.models.vit import Model

# Load the model
torch_model = Model.from_pretrained()

# Device
device = hub.Device("Samsung Galaxy S25")

# Trace model
input_shape = torch_model.get_input_spec()
sample_inputs = torch_model.sample_inputs()

pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])

# Compile model on a specific device
compile_job = hub.submit_compile_job(
    model=pt_model,
    device=device,
    input_specs=torch_model.get_input_spec(),
)

# Get target model to run on-device
target_model = compile_job.get_target_model()

Step 2: Performance profiling on cloud-hosted device

After compiling models from step 1. Models can be profiled model on-device using the target_model. Note that this scripts runs the model on a device automatically provisioned in the cloud. Once the job is submitted, you can navigate to a provided job URL to view a variety of on-device performance metrics.

profile_job = hub.submit_profile_job(
    model=target_model,
    device=device,
)

Step 3: Verify on-device accuracy

To verify the accuracy of the model on-device, you can run on-device inference on sample input data on the same cloud hosted device.

input_data = torch_model.sample_inputs()
inference_job = hub.submit_inference_job(
    model=target_model,
    device=device,
    inputs=input_data,
)
    on_device_output = inference_job.download_output_data()

With the output of the model, you can compute like PSNR, relative errors or spot check the output with expected output.

Note: This on-device profiling and inference requires access to Qualcomm® AI Hub. Sign up for access.

Run demo on a cloud-hosted device

You can also run the demo on-device.

python -m qai_hub_models.models.vit.demo --eval-mode on-device

NOTE: If you want running in a Jupyter Notebook or Google Colab like environment, please add the following to your cell (instead of the above).

%run -m qai_hub_models.models.vit.demo -- --eval-mode on-device

Deploying compiled model to Android

The models can be deployed using multiple runtimes:

  • TensorFlow Lite (.tflite export): This tutorial provides a guide to deploy the .tflite model in an Android application.

  • QNN (.so export ): This sample app provides instructions on how to use the .so shared library in an Android application.

View on Qualcomm® AI Hub

Get more details on VIT's performance across various devices here. Explore all available models on Qualcomm® AI Hub

License

  • The license for the original implementation of VIT can be found here.
  • The license for the compiled assets for on-device deployment can be found here

References

Community

Downloads last month
355
Inference Providers NEW
Image Classification
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support