Hugging Face's logo

qualcomm
/
Beit

Image Classification
PyTorch
backbone
android
Model card Files
xet
 Community

Beit: Optimized for Qualcomm Devices

Beit 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 Beit 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 Beit 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 Beit 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): 351 MB
  • Number of parameters: 92.0M

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
Beit ONNX float Snapdragon® X2 Elite 7.451 ms 2 - 2 MB NPU
Beit ONNX float Snapdragon® X Elite 15.395 ms 185 - 185 MB NPU
Beit ONNX float Snapdragon® 8 Gen 3 Mobile 10.475 ms 0 - 449 MB NPU
Beit ONNX float Snapdragon® 8 Gen 1 Mobile 17.996 ms 1 - 424 MB NPU
Beit ONNX float Qualcomm® Dragonwing™ IQ-8275 18.068 ms 0 - 4 MB NPU
Beit ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 14.828 ms 0 - 224 MB NPU
Beit ONNX float Qualcomm® QCS8450 17.996 ms 1 - 424 MB NPU
Beit ONNX float Qualcomm® Dragonwing™ IQ-9075 17.684 ms 1 - 4 MB NPU
Beit ONNX float Qualcomm® Dragonwing™ IQ-X7181 15.395 ms 185 - 185 MB NPU
Beit ONNX float Qualcomm® Dragonwing™ Q-8750 8.547 ms 1 - 307 MB NPU
Beit ONNX float Snapdragon® 8 Elite Mobile 8.547 ms 1 - 307 MB NPU
Beit ONNX float Snapdragon® 8 Elite Gen 5 Mobile 7.166 ms 0 - 305 MB NPU
Beit ONNX w8a16 Snapdragon® X2 Elite 3.338 ms 1 - 1 MB NPU
Beit ONNX w8a16 Snapdragon® X Elite 6.903 ms 96 - 96 MB NPU
Beit ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 4.605 ms 0 - 425 MB NPU
Beit ONNX w8a16 Snapdragon® 8 Gen 1 Mobile 11.13 ms 0 - 423 MB NPU
Beit ONNX w8a16 Qualcomm® Dragonwing™ QCS6490 36.083 ms 0 - 3 MB NPU
Beit ONNX w8a16 Qualcomm® Dragonwing™ IQ-8275 6.264 ms 0 - 4 MB NPU
Beit ONNX w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 6.683 ms 0 - 4 MB NPU
Beit ONNX w8a16 Qualcomm® QCS8450 11.13 ms 0 - 423 MB NPU
Beit ONNX w8a16 Qualcomm® Dragonwing™ IQ-9075 6.631 ms 0 - 3 MB NPU
Beit ONNX w8a16 Qualcomm® Dragonwing™ IQ-X7181 6.903 ms 96 - 96 MB NPU
Beit ONNX w8a16 Qualcomm® Dragonwing™ Q-6690 115.35 ms 0 - 441 MB NPU
Beit ONNX w8a16 Qualcomm® Dragonwing™ Q-7790 9.044 ms 0 - 411 MB NPU
Beit ONNX w8a16 Qualcomm® Dragonwing™ Q-8750 3.585 ms 0 - 362 MB NPU
Beit ONNX w8a16 Snapdragon® 8 Elite Mobile 3.585 ms 0 - 362 MB NPU
Beit ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile 2.399 ms 0 - 257 MB NPU
Beit ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 9.044 ms 0 - 411 MB NPU
Beit QNN_DLC float Snapdragon® X2 Elite 7.856 ms 1 - 1 MB NPU
Beit QNN_DLC float Snapdragon® X Elite 15.237 ms 1 - 1 MB NPU
Beit QNN_DLC float Snapdragon® 8 Gen 3 Mobile 10.518 ms 1 - 396 MB NPU
Beit QNN_DLC float Snapdragon® 8 Gen 1 Mobile 17.494 ms 1 - 384 MB NPU
Beit QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 17.868 ms 1 - 4 MB NPU
Beit QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 47.959 ms 1 - 296 MB NPU
Beit QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 14.363 ms 1 - 2 MB NPU
Beit QNN_DLC float Qualcomm® SA8775P 17.182 ms 1 - 296 MB NPU
Beit QNN_DLC float Qualcomm® SA8650P 17.182 ms 1 - 296 MB NPU
Beit QNN_DLC float Qualcomm® SA8255P 17.182 ms 1 - 296 MB NPU
Beit QNN_DLC float Qualcomm® QCS8450 17.494 ms 1 - 384 MB NPU
Beit QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 17.393 ms 3 - 5 MB NPU
Beit QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 15.237 ms 1 - 1 MB NPU
Beit QNN_DLC float Qualcomm® Dragonwing™ Q-8750 8.501 ms 1 - 300 MB NPU
Beit QNN_DLC float Qualcomm® SA7255P 47.959 ms 1 - 296 MB NPU
Beit QNN_DLC float Qualcomm® SA8295P 15.269 ms 1 - 289 MB NPU
Beit QNN_DLC float Snapdragon® 8 Elite Mobile 8.501 ms 1 - 300 MB NPU
Beit QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 7.185 ms 1 - 304 MB NPU
Beit QNN_DLC w8a16 Snapdragon® X2 Elite 3.819 ms 0 - 0 MB NPU
Beit QNN_DLC w8a16 Snapdragon® X Elite 7.346 ms 0 - 0 MB NPU
Beit QNN_DLC w8a16 Snapdragon® 8 Gen 3 Mobile 4.677 ms 0 - 412 MB NPU
Beit QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-8275 6.279 ms 0 - 3 MB NPU
Beit QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-8275 15.092 ms 0 - 351 MB NPU
Beit QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 6.781 ms 0 - 3 MB NPU
Beit QNN_DLC w8a16 Qualcomm® SA8775P 7.069 ms 0 - 351 MB NPU
Beit QNN_DLC w8a16 Qualcomm® SA8650P 7.069 ms 0 - 351 MB NPU
Beit QNN_DLC w8a16 Qualcomm® SA8255P 7.069 ms 0 - 351 MB NPU
Beit QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-9075 6.875 ms 0 - 2 MB NPU
Beit QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-X7181 7.346 ms 0 - 0 MB NPU
Beit QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-6690 66.219 ms 0 - 428 MB NPU
Beit QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-7790 8.695 ms 0 - 395 MB NPU
Beit QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-8750 3.55 ms 0 - 351 MB NPU
Beit QNN_DLC w8a16 Qualcomm® SA7255P 15.092 ms 0 - 351 MB NPU
Beit QNN_DLC w8a16 Snapdragon® 8 Elite Mobile 3.55 ms 0 - 351 MB NPU
Beit QNN_DLC w8a16 Snapdragon® 8 Elite Gen 5 Mobile 2.551 ms 0 - 246 MB NPU
Beit QNN_DLC w8a16 Snapdragon® 7 Gen 4 Mobile 8.695 ms 0 - 395 MB NPU
Beit TFLITE float Snapdragon® 8 Gen 3 Mobile 10.539 ms 0 - 405 MB NPU
Beit TFLITE float Snapdragon® 8 Gen 1 Mobile 17.577 ms 1 - 372 MB NPU
Beit TFLITE float Qualcomm® Dragonwing™ IQ-8275 17.878 ms 0 - 187 MB NPU
Beit TFLITE float Qualcomm® Dragonwing™ IQ-8275 48.032 ms 0 - 300 MB NPU
Beit TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 14.017 ms 0 - 3 MB NPU
Beit TFLITE float Qualcomm® SA8775P 17.188 ms 0 - 302 MB NPU
Beit TFLITE float Qualcomm® SA8650P 17.188 ms 0 - 302 MB NPU
Beit TFLITE float Qualcomm® SA8255P 17.188 ms 0 - 302 MB NPU
Beit TFLITE float Qualcomm® QCS8450 17.577 ms 1 - 372 MB NPU
Beit TFLITE float Qualcomm® Dragonwing™ IQ-9075 17.104 ms 0 - 186 MB NPU
Beit TFLITE float Qualcomm® Dragonwing™ Q-8750 8.548 ms 0 - 311 MB NPU
Beit TFLITE float Qualcomm® SA7255P 48.032 ms 0 - 300 MB NPU
Beit TFLITE float Qualcomm® SA8295P 15.341 ms 0 - 293 MB NPU
Beit TFLITE float Snapdragon® 8 Elite Mobile 8.548 ms 0 - 311 MB NPU
Beit TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 7.181 ms 0 - 313 MB NPU

License

  • The license for the original implementation of Beit can be found here.

References

Community

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
Image Classification
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Paper for qualcomm/Beit