Hugging Face's logo

qualcomm
/
EfficientFormer

Image Classification
PyTorch
bu_auto
android
Model card Files
xet
 Community

EfficientFormer: Optimized for Qualcomm Devices

EfficientFormer is a vision transformer model that can classify images from the Imagenet dataset.

This is based on the implementation of EfficientFormer 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 EfficientFormer 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 EfficientFormer on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.image_classification

Model Stats:

  • Input resolution: 224x224
  • Model checkpoint: efficientformer_l1_300d
  • Model size (float): 46.9 MB
  • Model size (w8a16): 12.2 MB
  • Number of parameters: 12.3M

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
EfficientFormer ONNX float Snapdragon® X2 Elite 0.563 ms 2 - 2 MB NPU
EfficientFormer ONNX float Snapdragon® X Elite 1.199 ms 25 - 25 MB NPU
EfficientFormer ONNX float Snapdragon® 8 Gen 3 Mobile 0.83 ms 0 - 77 MB NPU
EfficientFormer ONNX float Snapdragon® 8 Gen 1 Mobile 4.974 ms 1 - 80 MB NPU
EfficientFormer ONNX float Qualcomm® Dragonwing™ IQ-8275 1.591 ms 1 - 4 MB NPU
EfficientFormer ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.155 ms 0 - 29 MB NPU
EfficientFormer ONNX float Qualcomm® QCS8450 4.974 ms 1 - 80 MB NPU
EfficientFormer ONNX float Qualcomm® Dragonwing™ IQ-9075 1.598 ms 1 - 3 MB NPU
EfficientFormer ONNX float Qualcomm® Dragonwing™ IQ-X7181 1.199 ms 25 - 25 MB NPU
EfficientFormer ONNX float Qualcomm® Dragonwing™ Q-8750 0.647 ms 0 - 41 MB NPU
EfficientFormer ONNX float Snapdragon® 8 Elite Mobile 0.647 ms 0 - 41 MB NPU
EfficientFormer ONNX float Snapdragon® 8 Elite Gen 5 Mobile 0.57 ms 0 - 45 MB NPU
EfficientFormer ONNX w8a16 Snapdragon® X2 Elite 0.58 ms 1 - 1 MB NPU
EfficientFormer ONNX w8a16 Snapdragon® X Elite 1.263 ms 12 - 12 MB NPU
EfficientFormer ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 0.825 ms 0 - 84 MB NPU
EfficientFormer ONNX w8a16 Snapdragon® 8 Gen 1 Mobile 1.622 ms 0 - 86 MB NPU
EfficientFormer ONNX w8a16 Qualcomm® Dragonwing™ QCS6490 3.073 ms 0 - 3 MB NPU
EfficientFormer ONNX w8a16 Qualcomm® Dragonwing™ IQ-8275 1.241 ms 0 - 4 MB NPU
EfficientFormer ONNX w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.212 ms 0 - 16 MB NPU
EfficientFormer ONNX w8a16 Qualcomm® QCS8450 1.622 ms 0 - 86 MB NPU
EfficientFormer ONNX w8a16 Qualcomm® Dragonwing™ IQ-9075 1.545 ms 0 - 3 MB NPU
EfficientFormer ONNX w8a16 Qualcomm® Dragonwing™ IQ-X7181 1.263 ms 12 - 12 MB NPU
EfficientFormer ONNX w8a16 Qualcomm® Dragonwing™ Q-6690 5.92 ms 0 - 67 MB NPU
EfficientFormer ONNX w8a16 Qualcomm® Dragonwing™ Q-7790 1.345 ms 0 - 65 MB NPU
EfficientFormer ONNX w8a16 Qualcomm® Dragonwing™ Q-8750 0.571 ms 0 - 59 MB NPU
EfficientFormer ONNX w8a16 Snapdragon® 8 Elite Mobile 0.571 ms 0 - 59 MB NPU
EfficientFormer ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile 0.477 ms 0 - 65 MB NPU
EfficientFormer ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 1.345 ms 0 - 65 MB NPU
EfficientFormer QNN_DLC float Snapdragon® X2 Elite 0.856 ms 1 - 1 MB NPU
EfficientFormer QNN_DLC float Snapdragon® X Elite 1.692 ms 1 - 1 MB NPU
EfficientFormer QNN_DLC float Snapdragon® 8 Gen 3 Mobile 1.042 ms 0 - 77 MB NPU
EfficientFormer QNN_DLC float Snapdragon® 8 Gen 1 Mobile 5.581 ms 1 - 78 MB NPU
EfficientFormer QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 1.837 ms 1 - 4 MB NPU
EfficientFormer QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 4.909 ms 1 - 40 MB NPU
EfficientFormer QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.465 ms 1 - 3 MB NPU
EfficientFormer QNN_DLC float Qualcomm® SA8775P 2.063 ms 1 - 43 MB NPU
EfficientFormer QNN_DLC float Qualcomm® SA8650P 2.063 ms 1 - 43 MB NPU
EfficientFormer QNN_DLC float Qualcomm® SA8255P 2.063 ms 1 - 43 MB NPU
EfficientFormer QNN_DLC float Qualcomm® QCS8450 5.581 ms 1 - 78 MB NPU
EfficientFormer QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 1.884 ms 1 - 3 MB NPU
EfficientFormer QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 1.692 ms 1 - 1 MB NPU
EfficientFormer QNN_DLC float Qualcomm® Dragonwing™ Q-8750 0.788 ms 0 - 46 MB NPU
EfficientFormer QNN_DLC float Qualcomm® SA7255P 4.909 ms 1 - 40 MB NPU
EfficientFormer QNN_DLC float Qualcomm® SA8295P 4.009 ms 0 - 39 MB NPU
EfficientFormer QNN_DLC float Snapdragon® 8 Elite Mobile 0.788 ms 0 - 46 MB NPU
EfficientFormer QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 0.652 ms 1 - 44 MB NPU
EfficientFormer QNN_DLC w8a16 Snapdragon® X2 Elite 0.793 ms 0 - 0 MB NPU
EfficientFormer QNN_DLC w8a16 Snapdragon® X Elite 1.691 ms 0 - 0 MB NPU
EfficientFormer QNN_DLC w8a16 Snapdragon® 8 Gen 3 Mobile 1.0 ms 0 - 70 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-8275 1.441 ms 0 - 3 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-8275 3.08 ms 0 - 55 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.478 ms 0 - 2 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® SA8775P 1.772 ms 0 - 56 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® SA8650P 1.772 ms 0 - 56 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® SA8255P 1.772 ms 0 - 56 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-9075 1.612 ms 0 - 2 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-X7181 1.691 ms 0 - 0 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-6690 6.784 ms 0 - 176 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-7790 1.625 ms 0 - 58 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-8750 0.672 ms 0 - 49 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® SA7255P 3.08 ms 0 - 55 MB NPU
EfficientFormer QNN_DLC w8a16 Snapdragon® 8 Elite Mobile 0.672 ms 0 - 49 MB NPU
EfficientFormer QNN_DLC w8a16 Snapdragon® 8 Elite Gen 5 Mobile 0.571 ms 0 - 60 MB NPU
EfficientFormer QNN_DLC w8a16 Snapdragon® 7 Gen 4 Mobile 1.625 ms 0 - 58 MB NPU
EfficientFormer TFLITE float Snapdragon® 8 Gen 3 Mobile 1.041 ms 0 - 95 MB NPU
EfficientFormer TFLITE float Snapdragon® 8 Gen 1 Mobile 5.538 ms 0 - 96 MB NPU
EfficientFormer TFLITE float Qualcomm® Dragonwing™ IQ-8275 1.822 ms 0 - 28 MB NPU
EfficientFormer TFLITE float Qualcomm® Dragonwing™ IQ-8275 4.866 ms 0 - 53 MB NPU
EfficientFormer TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.485 ms 0 - 4 MB NPU
EfficientFormer TFLITE float Qualcomm® SA8775P 2.06 ms 0 - 55 MB NPU
EfficientFormer TFLITE float Qualcomm® SA8650P 2.06 ms 0 - 55 MB NPU
EfficientFormer TFLITE float Qualcomm® SA8255P 2.06 ms 0 - 55 MB NPU
EfficientFormer TFLITE float Qualcomm® QCS8450 5.538 ms 0 - 96 MB NPU
EfficientFormer TFLITE float Qualcomm® Dragonwing™ IQ-9075 1.868 ms 0 - 27 MB NPU
EfficientFormer TFLITE float Qualcomm® Dragonwing™ Q-8750 0.765 ms 0 - 49 MB NPU
EfficientFormer TFLITE float Qualcomm® SA7255P 4.866 ms 0 - 53 MB NPU
EfficientFormer TFLITE float Qualcomm® SA8295P 3.969 ms 0 - 48 MB NPU
EfficientFormer TFLITE float Snapdragon® 8 Elite Mobile 0.765 ms 0 - 49 MB NPU
EfficientFormer TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 0.653 ms 0 - 52 MB NPU

License

  • The license for the original implementation of EfficientFormer 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/EfficientFormer