---
library_name: pytorch
license: other
tags:
- bu_auto
- android
pipeline_tag: image-classification

---

![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/web-assets/model_demo.png)

# 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](https://github.com/snap-research/EfficientFormer).
This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.58.0/src/qai_hub_models/models/efficientformer) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).

Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) 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 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.58.0/efficientformer-onnx-float.zip)
| ONNX | w8a16 | Universal | QAIRT 2.45, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.58.0/efficientformer-onnx-w8a16.zip)
| QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.58.0/efficientformer-qnn_dlc-float.zip)
| QNN_DLC | w8a16 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.58.0/efficientformer-qnn_dlc-w8a16.zip)
| TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.58.0/efficientformer-tflite-float.zip)

For more device-specific assets and performance metrics, visit **[EfficientFormer on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/efficientformer)**.


### Option 2: Export with Custom Configurations

Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.58.0/src/qai_hub_models/models/efficientformer) 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](https://github.com/qualcomm/ai-hub-models/blob/v0.58.0/src/qai_hub_models/models/efficientformer) for usage instructions.

## Model Details

**Model Type:** Model_use_case.image_classification

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

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| EfficientFormer | ONNX | float | Snapdragon® X2 Elite | 0.635 ms | 2 - 2 MB | NPU
| EfficientFormer | ONNX | float | Snapdragon® X Elite | 1.376 ms | 24 - 24 MB | NPU
| EfficientFormer | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.915 ms | 0 - 79 MB | NPU
| EfficientFormer | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 5.095 ms | 1 - 83 MB | NPU
| EfficientFormer | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.324 ms | 0 - 39 MB | NPU
| EfficientFormer | ONNX | float | Qualcomm® QCS8450 | 5.095 ms | 1 - 83 MB | NPU
| EfficientFormer | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 1.759 ms | 1 - 3 MB | NPU
| EfficientFormer | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.612 ms | 0 - 45 MB | NPU
| EfficientFormer | ONNX | float | Snapdragon® 8 Elite Mobile | 0.69 ms | 0 - 44 MB | NPU
| EfficientFormer | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 0.69 ms | 0 - 44 MB | NPU
| EfficientFormer | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.376 ms | 24 - 24 MB | NPU
| EfficientFormer | ONNX | w8a16 | Snapdragon® X2 Elite | 0.567 ms | 1 - 1 MB | NPU
| EfficientFormer | ONNX | w8a16 | Snapdragon® X Elite | 1.378 ms | 12 - 12 MB | NPU
| EfficientFormer | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 0.898 ms | 0 - 95 MB | NPU
| EfficientFormer | ONNX | w8a16 | Snapdragon® 8 Gen 1 Mobile | 2.216 ms | 0 - 87 MB | NPU
| EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 5.752 ms | 1 - 3 MB | NPU
| EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.319 ms | 0 - 3 MB | NPU
| EfficientFormer | ONNX | w8a16 | Qualcomm® QCS8450 | 2.216 ms | 0 - 87 MB | NPU
| EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 1.573 ms | 0 - 3 MB | NPU
| EfficientFormer | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 1.469 ms | 0 - 70 MB | NPU
| EfficientFormer | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.532 ms | 0 - 72 MB | NPU
| EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 6.234 ms | 0 - 187 MB | NPU
| EfficientFormer | ONNX | w8a16 | Snapdragon® 8 Elite Mobile | 0.609 ms | 0 - 61 MB | NPU
| EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 1.469 ms | 0 - 70 MB | NPU
| EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 0.609 ms | 0 - 61 MB | NPU
| EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 1.378 ms | 12 - 12 MB | NPU
| EfficientFormer | QNN_DLC | float | Snapdragon® X2 Elite | 0.92 ms | 1 - 1 MB | NPU
| EfficientFormer | QNN_DLC | float | Snapdragon® X Elite | 1.675 ms | 1 - 1 MB | NPU
| EfficientFormer | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 1.036 ms | 0 - 75 MB | NPU
| EfficientFormer | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 5.577 ms | 0 - 77 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® QCS8275 | 4.896 ms | 0 - 40 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.479 ms | 1 - 2 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® SA8775P | 2.068 ms | 1 - 43 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® SA8650P | 2.068 ms | 1 - 43 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® SA8255P | 2.068 ms | 1 - 43 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® QCS8450 | 5.577 ms | 0 - 77 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 2.108 ms | 1 - 3 MB | NPU
| EfficientFormer | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.65 ms | 1 - 45 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® SA7255P | 4.896 ms | 0 - 40 MB | NPU
| EfficientFormer | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 0.769 ms | 0 - 44 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® SA8295P | 3.989 ms | 1 - 39 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 0.769 ms | 0 - 44 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.675 ms | 1 - 1 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 0.885 ms | 0 - 0 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Snapdragon® X Elite | 1.781 ms | 0 - 0 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 1.075 ms | 0 - 80 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® QCS8275 | 3.241 ms | 0 - 58 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.563 ms | 0 - 2 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® SA8775P | 1.892 ms | 0 - 59 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® SA8650P | 1.892 ms | 0 - 59 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® SA8255P | 1.892 ms | 0 - 59 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 1.735 ms | 2 - 4 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 1.692 ms | 0 - 62 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.615 ms | 0 - 65 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 6.987 ms | 0 - 178 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® SA7255P | 3.241 ms | 0 - 58 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Snapdragon® 8 Elite Mobile | 0.705 ms | 0 - 52 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 1.692 ms | 0 - 62 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 0.705 ms | 0 - 52 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 1.781 ms | 0 - 0 MB | NPU
| EfficientFormer | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 1.031 ms | 0 - 88 MB | NPU
| EfficientFormer | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 5.541 ms | 0 - 96 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® QCS8275 | 4.852 ms | 0 - 52 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.469 ms | 0 - 29 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® SA8775P | 2.066 ms | 0 - 53 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® SA8650P | 2.066 ms | 0 - 53 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® SA8255P | 2.066 ms | 0 - 53 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® QCS8450 | 5.541 ms | 0 - 96 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 1.886 ms | 0 - 27 MB | NPU
| EfficientFormer | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.655 ms | 0 - 53 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® SA7255P | 4.852 ms | 0 - 52 MB | NPU
| EfficientFormer | TFLITE | float | Snapdragon® 8 Elite Mobile | 0.779 ms | 0 - 56 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® SA8295P | 4.018 ms | 0 - 47 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 0.779 ms | 0 - 56 MB | NPU

## License
* The license for the original implementation of EfficientFormer can be found
  [here](https://github.com/snap-research/EfficientFormer?tab=License-1-ov-file#readme).

## References
* [Rethinking Vision Transformers for MobileNet Size and Speed](https://arxiv.org/abs/2212.08059)
* [Source Model Implementation](https://github.com/snap-research/EfficientFormer)

## Community
* Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
* For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).