Hugging Peers

EfficientFormer

☆ Save
by Qualcomm-AI-Hub image-classification apache-2.0
0 downloads 0 likes on ModelScope updated

Model card

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 0.564 ms 2 - 2 MB NPU
EfficientFormer ONNX float Snapdragon® X Elite 1.204 ms 24 - 24 MB NPU
EfficientFormer ONNX float Snapdragon® 8 Gen 3 Mobile 0.833 ms 0 - 76 MB NPU
EfficientFormer ONNX float Snapdragon® 8 Gen 1 Mobile 4.975 ms 1 - 80 MB NPU
EfficientFormer ONNX float Qualcomm® Dragonwing™ IQ-8275 1.594 ms 1 - 4 MB NPU
EfficientFormer ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.165 ms 0 - 133 MB NPU
EfficientFormer ONNX float Qualcomm® QCS8450 4.975 ms 1 - 80 MB NPU
EfficientFormer ONNX float Qualcomm® Dragonwing™ IQ-9075 1.583 ms 1 - 3 MB NPU
EfficientFormer ONNX float Qualcomm® Dragonwing™ IQ-X7181 1.204 ms 24 - 24 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 0.571 ms 0 - 46 MB NPU
EfficientFormer ONNX w8a16 Snapdragon® X2 Elite 0.52 ms 1 - 1 MB NPU
EfficientFormer ONNX w8a16 Snapdragon® X Elite 1.261 ms 13 - 13 MB NPU
EfficientFormer ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 0.821 ms 0 - 84 MB NPU
EfficientFormer ONNX w8a16 Snapdragon® 8 Gen 1 Mobile 1.613 ms 0 - 86 MB NPU
EfficientFormer ONNX w8a16 Qualcomm® Dragonwing™ QCS6490 3.327 ms 0 - 3 MB NPU
EfficientFormer ONNX w8a16 Qualcomm® Dragonwing™ IQ-8275 1.231 ms 0 - 4 MB NPU
EfficientFormer ONNX w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.213 ms 0 - 130 MB NPU
EfficientFormer ONNX w8a16 Qualcomm® QCS8450 1.613 ms 0 - 86 MB NPU
EfficientFormer ONNX w8a16 Qualcomm® Dragonwing™ IQ-9075 1.376 ms 0 - 3 MB NPU
EfficientFormer ONNX w8a16 Qualcomm® Dragonwing™ IQ-X7181 1.261 ms 13 - 13 MB NPU
EfficientFormer ONNX w8a16 Qualcomm® Dragonwing™ Q-6690 5.924 ms 0 - 67 MB NPU
EfficientFormer ONNX w8a16 Qualcomm® Dragonwing™ Q-7790 1.361 ms 0 - 65 MB NPU
EfficientFormer ONNX w8a16 Qualcomm® Dragonwing™ Q-8750 0.575 ms 0 - 59 MB NPU
EfficientFormer ONNX w8a16 Snapdragon® 8 Elite Mobile 0.575 ms 0 - 59 MB NPU
EfficientFormer ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile 0.486 ms 0 - 65 MB NPU
EfficientFormer ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 1.361 ms 0 - 65 MB NPU
EfficientFormer QNN_DLC float Snapdragon® X2 Elite 0.904 ms 1 - 1 MB NPU
EfficientFormer QNN_DLC float Snapdragon® X Elite 1.676 ms 1 - 1 MB NPU
EfficientFormer QNN_DLC float Snapdragon® 8 Gen 3 Mobile 1.029 ms 0 - 76 MB NPU
EfficientFormer QNN_DLC float Snapdragon® 8 Gen 1 Mobile 5.588 ms 0 - 78 MB NPU
EfficientFormer QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 1.837 ms 1 - 4 MB NPU
EfficientFormer QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.473 ms 1 - 2 MB NPU
EfficientFormer QNN_DLC float Qualcomm® SA8775P 2.046 ms 1 - 43 MB NPU
EfficientFormer QNN_DLC float Qualcomm® SA8650P 2.046 ms 1 - 43 MB NPU
EfficientFormer QNN_DLC float Qualcomm® SA8255P 2.046 ms 1 - 43 MB NPU
EfficientFormer QNN_DLC float Qualcomm® QCS8450 5.588 ms 0 - 78 MB NPU
EfficientFormer QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 1.88 ms 3 - 5 MB NPU
EfficientFormer QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 1.676 ms 1 - 1 MB NPU
EfficientFormer QNN_DLC float Qualcomm® Dragonwing™ Q-8750 0.774 ms 0 - 40 MB NPU
EfficientFormer QNN_DLC float Qualcomm® SA7255P 4.897 ms 1 - 40 MB NPU
EfficientFormer QNN_DLC float Qualcomm® SA8295P 4.007 ms 0 - 39 MB NPU
EfficientFormer QNN_DLC float Snapdragon® 8 Elite Mobile 0.774 ms 0 - 40 MB NPU
EfficientFormer QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 0.655 ms 1 - 43 MB NPU
EfficientFormer QNN_DLC w8a16 Snapdragon® X2 Elite 0.818 ms 0 - 0 MB NPU
EfficientFormer QNN_DLC w8a16 Snapdragon® X Elite 1.67 ms 0 - 0 MB NPU
EfficientFormer QNN_DLC w8a16 Snapdragon® 8 Gen 3 Mobile 1.009 ms 0 - 75 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-8275 1.433 ms 0 - 3 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.473 ms 0 - 89 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® SA8775P 1.781 ms 0 - 56 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® SA8650P 1.781 ms 0 - 56 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® SA8255P 1.781 ms 0 - 56 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-9075 1.628 ms 0 - 2 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-X7181 1.67 ms 0 - 0 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-6690 6.859 ms 0 - 176 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-7790 1.621 ms 0 - 58 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-8750 0.674 ms 0 - 49 MB NPU
EfficientFormer QNN_DLC w8a16 Qualcomm® SA7255P 3.055 ms 0 - 53 MB NPU
EfficientFormer QNN_DLC w8a16 Snapdragon® 8 Elite Mobile 0.674 ms 0 - 49 MB NPU
EfficientFormer QNN_DLC w8a16 Snapdragon® 8 Elite Gen 5 Mobile 0.564 ms 0 - 59 MB NPU
EfficientFormer QNN_DLC w8a16 Snapdragon® 7 Gen 4 Mobile 1.621 ms 0 - 58 MB NPU
EfficientFormer TFLITE float Snapdragon® 8 Gen 3 Mobile 1.034 ms 0 - 96 MB NPU
EfficientFormer TFLITE float Snapdragon® 8 Gen 1 Mobile 5.553 ms 0 - 94 MB NPU
EfficientFormer TFLITE float Qualcomm® Dragonwing™ IQ-8275 1.825 ms 0 - 28 MB NPU
EfficientFormer TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.493 ms 0 - 2 MB NPU
EfficientFormer TFLITE float Qualcomm® SA8775P 2.056 ms 0 - 55 MB NPU
EfficientFormer TFLITE float Qualcomm® SA8650P 2.056 ms 0 - 55 MB NPU
EfficientFormer TFLITE float Qualcomm® SA8255P 2.056 ms 0 - 55 MB NPU
EfficientFormer TFLITE float Qualcomm® QCS8450 5.553 ms 0 - 94 MB NPU
EfficientFormer TFLITE float Qualcomm® Dragonwing™ IQ-9075 1.876 ms 0 - 27 MB NPU
EfficientFormer TFLITE float Qualcomm® Dragonwing™ Q-8750 0.768 ms 0 - 49 MB NPU
EfficientFormer TFLITE float Qualcomm® SA7255P 4.838 ms 0 - 52 MB NPU
EfficientFormer TFLITE float Qualcomm® SA8295P 3.968 ms 0 - 48 MB NPU
EfficientFormer TFLITE float Snapdragon® 8 Elite Mobile 0.768 ms 0 - 49 MB NPU
EfficientFormer TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 0.657 ms 0 - 50 MB NPU

License

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

References

Community