Hugging Face's logo

qualcomm
/
GPUNet

Image Classification
PyTorch
LiteRT
backbone
android
Model card Files
xet
Community

GPUNet: Optimized for Mobile Deployment

Imagenet classifier and general purpose backbone

GPUNet 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 GPUNet found here.

This repository provides scripts to run GPUNet 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: 10.49M
    • Model size (float): 45.28MB
    • Model size (w8a8): 21.3MB
Model Precision Device Chipset Target Runtime Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit Target Model
GPUNet float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 4.759 ms 0 - 51 MB NPU GPUNet.tflite
GPUNet float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN_DLC 4.666 ms 1 - 24 MB NPU GPUNet.dlc
GPUNet float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 1.798 ms 0 - 65 MB NPU GPUNet.tflite
GPUNet float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN_DLC 2.252 ms 1 - 35 MB NPU GPUNet.dlc
GPUNet float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 1.244 ms 0 - 179 MB NPU GPUNet.tflite
GPUNet float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN_DLC 1.243 ms 0 - 83 MB NPU GPUNet.dlc
GPUNet float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) ONNX 1.248 ms 0 - 78 MB NPU GPUNet.onnx.zip
GPUNet float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 7.129 ms 0 - 51 MB NPU GPUNet.tflite
GPUNet float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN_DLC 1.708 ms 1 - 24 MB NPU GPUNet.dlc
GPUNet float SA7255P ADP Qualcomm® SA7255P TFLITE 4.759 ms 0 - 51 MB NPU GPUNet.tflite
GPUNet float SA7255P ADP Qualcomm® SA7255P QNN_DLC 4.666 ms 1 - 24 MB NPU GPUNet.dlc
GPUNet float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 1.237 ms 0 - 178 MB NPU GPUNet.tflite
GPUNet float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN_DLC 1.243 ms 0 - 88 MB NPU GPUNet.dlc
GPUNet float SA8295P ADP Qualcomm® SA8295P TFLITE 2.221 ms 0 - 56 MB NPU GPUNet.tflite
GPUNet float SA8295P ADP Qualcomm® SA8295P QNN_DLC 2.22 ms 1 - 31 MB NPU GPUNet.dlc
GPUNet float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 1.242 ms 0 - 179 MB NPU GPUNet.tflite
GPUNet float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN_DLC 1.245 ms 0 - 80 MB NPU GPUNet.dlc
GPUNet float SA8775P ADP Qualcomm® SA8775P TFLITE 7.129 ms 0 - 51 MB NPU GPUNet.tflite
GPUNet float SA8775P ADP Qualcomm® SA8775P QNN_DLC 1.708 ms 1 - 24 MB NPU GPUNet.dlc
GPUNet float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 0.888 ms 0 - 67 MB NPU GPUNet.tflite
GPUNet float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN_DLC 0.899 ms 1 - 37 MB NPU GPUNet.dlc
GPUNet float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 0.899 ms 0 - 36 MB NPU GPUNet.onnx.zip
GPUNet float Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile TFLITE 0.693 ms 0 - 58 MB NPU GPUNet.tflite
GPUNet float Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile QNN_DLC 0.697 ms 1 - 31 MB NPU GPUNet.dlc
GPUNet float Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile ONNX 0.707 ms 0 - 27 MB NPU GPUNet.onnx.zip
GPUNet float Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile TFLITE 0.583 ms 0 - 59 MB NPU GPUNet.tflite
GPUNet float Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile QNN_DLC 0.568 ms 1 - 32 MB NPU GPUNet.dlc
GPUNet float Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile ONNX 0.633 ms 1 - 28 MB NPU GPUNet.onnx.zip
GPUNet float Snapdragon X Elite CRD Snapdragon® X Elite QNN_DLC 1.345 ms 106 - 106 MB NPU GPUNet.dlc
GPUNet float Snapdragon X Elite CRD Snapdragon® X Elite ONNX 1.107 ms 24 - 24 MB NPU GPUNet.onnx.zip
GPUNet w8a16 Dragonwing RB3 Gen 2 Vision Kit Qualcomm® QCS6490 QNN_DLC 3.034 ms 0 - 100 MB NPU GPUNet.dlc
GPUNet w8a16 Dragonwing RB3 Gen 2 Vision Kit Qualcomm® QCS6490 ONNX 101.013 ms 21 - 36 MB CPU GPUNet.onnx.zip
GPUNet w8a16 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN_DLC 2.499 ms 0 - 31 MB NPU GPUNet.dlc
GPUNet w8a16 QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN_DLC 1.43 ms 0 - 43 MB NPU GPUNet.dlc
GPUNet w8a16 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN_DLC 1.069 ms 0 - 57 MB NPU GPUNet.dlc
GPUNet w8a16 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) ONNX 1.022 ms 0 - 58 MB NPU GPUNet.onnx.zip
GPUNet w8a16 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN_DLC 1.292 ms 0 - 31 MB NPU GPUNet.dlc
GPUNet w8a16 RB5 (Proxy) Qualcomm® QCS8250 (Proxy) ONNX 45.694 ms 21 - 38 MB CPU GPUNet.onnx.zip
GPUNet w8a16 SA7255P ADP Qualcomm® SA7255P QNN_DLC 2.499 ms 0 - 31 MB NPU GPUNet.dlc
GPUNet w8a16 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN_DLC 1.074 ms 0 - 57 MB NPU GPUNet.dlc
GPUNet w8a16 SA8295P ADP Qualcomm® SA8295P QNN_DLC 1.633 ms 0 - 37 MB NPU GPUNet.dlc
GPUNet w8a16 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN_DLC 1.07 ms 0 - 58 MB NPU GPUNet.dlc
GPUNet w8a16 SA8775P ADP Qualcomm® SA8775P QNN_DLC 1.292 ms 0 - 31 MB NPU GPUNet.dlc
GPUNet w8a16 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN_DLC 0.77 ms 0 - 45 MB NPU GPUNet.dlc
GPUNet w8a16 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 0.714 ms 0 - 40 MB NPU GPUNet.onnx.zip
GPUNet w8a16 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile QNN_DLC 0.519 ms 0 - 37 MB NPU GPUNet.dlc
GPUNet w8a16 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile ONNX 0.535 ms 0 - 35 MB NPU GPUNet.onnx.zip
GPUNet w8a16 Snapdragon 7 Gen 4 QRD Snapdragon® 7 Gen 4 Mobile QNN_DLC 1.303 ms 0 - 44 MB NPU GPUNet.dlc
GPUNet w8a16 Snapdragon 7 Gen 4 QRD Snapdragon® 7 Gen 4 Mobile ONNX 51.681 ms 29 - 45 MB CPU GPUNet.onnx.zip
GPUNet w8a16 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile QNN_DLC 0.437 ms 0 - 37 MB NPU GPUNet.dlc
GPUNet w8a16 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile ONNX 0.495 ms 0 - 33 MB NPU GPUNet.onnx.zip
GPUNet w8a16 Snapdragon X Elite CRD Snapdragon® X Elite QNN_DLC 1.227 ms 57 - 57 MB NPU GPUNet.dlc
GPUNet w8a16 Snapdragon X Elite CRD Snapdragon® X Elite ONNX 0.971 ms 12 - 12 MB NPU GPUNet.onnx.zip
GPUNet w8a8 Dragonwing RB3 Gen 2 Vision Kit Qualcomm® QCS6490 TFLITE 1.645 ms 0 - 15 MB NPU GPUNet.tflite
GPUNet w8a8 Dragonwing RB3 Gen 2 Vision Kit Qualcomm® QCS6490 QNN_DLC 1.773 ms 0 - 100 MB NPU GPUNet.dlc
GPUNet w8a8 Dragonwing RB3 Gen 2 Vision Kit Qualcomm® QCS6490 ONNX 16.81 ms 5 - 18 MB CPU GPUNet.onnx.zip
GPUNet w8a8 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 1.118 ms 0 - 30 MB NPU GPUNet.tflite
GPUNet w8a8 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN_DLC 1.406 ms 0 - 30 MB NPU GPUNet.dlc
GPUNet w8a8 QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 0.575 ms 0 - 44 MB NPU GPUNet.tflite
GPUNet w8a8 QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN_DLC 0.871 ms 0 - 42 MB NPU GPUNet.dlc
GPUNet w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 0.434 ms 0 - 62 MB NPU GPUNet.tflite
GPUNet w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN_DLC 0.623 ms 0 - 62 MB NPU GPUNet.dlc
GPUNet w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) ONNX 0.818 ms 0 - 34 MB NPU GPUNet.onnx.zip
GPUNet w8a8 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 0.635 ms 0 - 30 MB NPU GPUNet.tflite
GPUNet w8a8 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN_DLC 0.801 ms 0 - 30 MB NPU GPUNet.dlc
GPUNet w8a8 RB5 (Proxy) Qualcomm® QCS8250 (Proxy) TFLITE 8.105 ms 0 - 3 MB NPU GPUNet.tflite
GPUNet w8a8 RB5 (Proxy) Qualcomm® QCS8250 (Proxy) ONNX 9.809 ms 3 - 18 MB CPU GPUNet.onnx.zip
GPUNet w8a8 SA7255P ADP Qualcomm® SA7255P TFLITE 1.118 ms 0 - 30 MB NPU GPUNet.tflite
GPUNet w8a8 SA7255P ADP Qualcomm® SA7255P QNN_DLC 1.406 ms 0 - 30 MB NPU GPUNet.dlc
GPUNet w8a8 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 0.446 ms 0 - 57 MB NPU GPUNet.tflite
GPUNet w8a8 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN_DLC 0.613 ms 0 - 62 MB NPU GPUNet.dlc
GPUNet w8a8 SA8295P ADP Qualcomm® SA8295P TFLITE 0.864 ms 0 - 36 MB NPU GPUNet.tflite
GPUNet w8a8 SA8295P ADP Qualcomm® SA8295P QNN_DLC 1.054 ms 0 - 37 MB NPU GPUNet.dlc
GPUNet w8a8 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 0.45 ms 0 - 63 MB NPU GPUNet.tflite
GPUNet w8a8 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN_DLC 0.623 ms 0 - 61 MB NPU GPUNet.dlc
GPUNet w8a8 SA8775P ADP Qualcomm® SA8775P TFLITE 0.635 ms 0 - 30 MB NPU GPUNet.tflite
GPUNet w8a8 SA8775P ADP Qualcomm® SA8775P QNN_DLC 0.801 ms 0 - 30 MB NPU GPUNet.dlc
GPUNet w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 0.343 ms 0 - 48 MB NPU GPUNet.tflite
GPUNet w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN_DLC 0.462 ms 0 - 42 MB NPU GPUNet.dlc
GPUNet w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 0.562 ms 0 - 44 MB NPU GPUNet.onnx.zip
GPUNet w8a8 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile TFLITE 0.268 ms 0 - 30 MB NPU GPUNet.tflite
GPUNet w8a8 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile QNN_DLC 0.342 ms 0 - 34 MB NPU GPUNet.dlc
GPUNet w8a8 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile ONNX 0.502 ms 0 - 36 MB NPU GPUNet.onnx.zip
GPUNet w8a8 Snapdragon 7 Gen 4 QRD Snapdragon® 7 Gen 4 Mobile TFLITE 0.64 ms 0 - 39 MB NPU GPUNet.tflite
GPUNet w8a8 Snapdragon 7 Gen 4 QRD Snapdragon® 7 Gen 4 Mobile QNN_DLC 0.798 ms 0 - 43 MB NPU GPUNet.dlc
GPUNet w8a8 Snapdragon 7 Gen 4 QRD Snapdragon® 7 Gen 4 Mobile ONNX 9.661 ms 10 - 26 MB CPU GPUNet.onnx.zip
GPUNet w8a8 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile TFLITE 0.254 ms 0 - 33 MB NPU GPUNet.tflite
GPUNet w8a8 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile QNN_DLC 0.283 ms 0 - 35 MB NPU GPUNet.dlc
GPUNet w8a8 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile ONNX 0.473 ms 0 - 34 MB NPU GPUNet.onnx.zip
GPUNet w8a8 Snapdragon X Elite CRD Snapdragon® X Elite QNN_DLC 0.721 ms 64 - 64 MB NPU GPUNet.dlc
GPUNet w8a8 Snapdragon X Elite CRD Snapdragon® X Elite ONNX 0.716 ms 12 - 12 MB NPU GPUNet.onnx.zip

Installation

Install the package via pip:

pip install qai-hub-models

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

Sign-in to Qualcomm® AI Hub Workbench 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.gpunet.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.gpunet.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.gpunet.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.gpunet 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 Workbench. Sign up for access.

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 GPUNet's performance across various devices here. Explore all available models on Qualcomm® AI Hub

License

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

References

Community

Downloads last month
63
Inference Providers NEW
Image Classification
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support