---
library_name: pytorch
license: other
tags:
- android
pipeline_tag: image-to-text

---

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

# EasyOCR: Optimized for Qualcomm Devices

EasyOCR is a machine learning model that can recognize text in images. It supports 80+ supported languages and all popular writing scripts.

This is based on the implementation of EasyOCR found [here](https://github.com/JaidedAI/EasyOCR).
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/main/src/qai_hub_models/models/easyocr) 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 2.42, ONNX Runtime 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/easyocr/releases/v0.50.2/easyocr-onnx-float.zip)
| ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/easyocr/releases/v0.50.2/easyocr-onnx-w8a8.zip)
| QNN_DLC | float | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/easyocr/releases/v0.50.2/easyocr-qnn_dlc-float.zip)
| TFLITE | float | Universal | QAIRT 2.43, TFLite 2.19.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/easyocr/releases/v0.50.2/easyocr-tflite-float.zip)
| TFLITE | w8a8 | Universal | QAIRT 2.43, TFLite 2.19.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/easyocr/releases/v0.50.2/easyocr-tflite-w8a8.zip)

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


### Option 2: Export with Custom Configurations

Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/easyocr) 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 [EasyOCR on GitHub](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/easyocr) for usage instructions.

## Model Details

**Model Type:** Model_use_case.image_to_text

**Model Stats:**
- Model checkpoint: easyocr-small-stage1
- Input resolution: 608x800
- Number of parameters (detector): 20.8M
- Model size (detector) (float): 79.2 MB
- Number of parameters (recognizer): 3.84M
- Model size (recognizer) (float): 14.7 MB

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| detector | ONNX | float | Snapdragon® X2 Elite | 19.987 ms | 36 - 36 MB | NPU
| detector | ONNX | float | Snapdragon® X Elite | 38.228 ms | 35 - 35 MB | NPU
| detector | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 27.357 ms | 3 - 238 MB | NPU
| detector | ONNX | float | Qualcomm® QCS8550 (Proxy) | 37.121 ms | 6 - 8 MB | NPU
| detector | ONNX | float | Qualcomm® QCS9075 | 69.962 ms | 6 - 14 MB | NPU
| detector | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 21.926 ms | 3 - 171 MB | NPU
| detector | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 5.361 ms | 0 - 161 MB | NPU
| detector | ONNX | w8a8 | Snapdragon® X2 Elite | 26.382 ms | 20 - 20 MB | NPU
| detector | ONNX | w8a8 | Snapdragon® X Elite | 13.674 ms | 19 - 19 MB | NPU
| detector | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 9.749 ms | 2 - 217 MB | NPU
| detector | ONNX | w8a8 | Qualcomm® QCS6490 | 948.125 ms | 224 - 243 MB | CPU
| detector | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 13.061 ms | 1 - 198 MB | NPU
| detector | ONNX | w8a8 | Qualcomm® QCS9075 | 13.44 ms | 1 - 4 MB | NPU
| detector | ONNX | w8a8 | Qualcomm® QCM6690 | 927.919 ms | 233 - 241 MB | CPU
| detector | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 7.759 ms | 2 - 162 MB | NPU
| detector | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 920.173 ms | 233 - 241 MB | CPU
| detector | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 16.526 ms | 6 - 183 MB | NPU
| detector | QNN_DLC | float | Snapdragon® X2 Elite | 20.651 ms | 6 - 6 MB | NPU
| detector | QNN_DLC | float | Snapdragon® X Elite | 40.855 ms | 6 - 6 MB | NPU
| detector | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 28.629 ms | 6 - 248 MB | NPU
| detector | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 274.902 ms | 0 - 171 MB | NPU
| detector | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 39.213 ms | 6 - 7 MB | NPU
| detector | QNN_DLC | float | Qualcomm® SA8775P | 70.201 ms | 1 - 171 MB | NPU
| detector | QNN_DLC | float | Qualcomm® QCS9075 | 76.412 ms | 6 - 14 MB | NPU
| detector | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 80.478 ms | 6 - 262 MB | NPU
| detector | QNN_DLC | float | Qualcomm® SA7255P | 274.902 ms | 0 - 171 MB | NPU
| detector | QNN_DLC | float | Qualcomm® SA8295P | 77.621 ms | 0 - 183 MB | NPU
| detector | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 22.477 ms | 0 - 171 MB | NPU
| detector | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 16.467 ms | 1 - 184 MB | NPU
| detector | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 28.726 ms | 1 - 255 MB | NPU
| detector | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 275.045 ms | 1 - 179 MB | NPU
| detector | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 39.308 ms | 1 - 3 MB | NPU
| detector | TFLITE | float | Qualcomm® SA8775P | 70.269 ms | 1 - 178 MB | NPU
| detector | TFLITE | float | Qualcomm® QCS9075 | 75.468 ms | 0 - 49 MB | NPU
| detector | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 81.786 ms | 1 - 271 MB | NPU
| detector | TFLITE | float | Qualcomm® SA7255P | 275.045 ms | 1 - 179 MB | NPU
| detector | TFLITE | float | Qualcomm® SA8295P | 73.833 ms | 1 - 187 MB | NPU
| detector | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 22.512 ms | 1 - 180 MB | NPU
| detector | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 4.463 ms | 0 - 158 MB | NPU
| detector | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 7.153 ms | 0 - 213 MB | NPU
| detector | TFLITE | w8a8 | Qualcomm® QCS6490 | 51.307 ms | 0 - 24 MB | NPU
| detector | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 35.725 ms | 0 - 156 MB | NPU
| detector | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 9.506 ms | 0 - 3 MB | NPU
| detector | TFLITE | w8a8 | Qualcomm® SA8775P | 10.092 ms | 0 - 155 MB | NPU
| detector | TFLITE | w8a8 | Qualcomm® QCS9075 | 11.75 ms | 0 - 24 MB | NPU
| detector | TFLITE | w8a8 | Qualcomm® QCM6690 | 235.276 ms | 0 - 208 MB | NPU
| detector | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 14.57 ms | 0 - 212 MB | NPU
| detector | TFLITE | w8a8 | Qualcomm® SA7255P | 35.725 ms | 0 - 156 MB | NPU
| detector | TFLITE | w8a8 | Qualcomm® SA8295P | 18.939 ms | 0 - 153 MB | NPU
| detector | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 5.767 ms | 0 - 158 MB | NPU
| detector | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 18.942 ms | 0 - 176 MB | NPU
| recognizer | ONNX | float | Snapdragon® X2 Elite | 11.97 ms | 13 - 13 MB | NPU
| recognizer | ONNX | float | Snapdragon® X Elite | 16.235 ms | 11 - 11 MB | NPU
| recognizer | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 12.121 ms | 0 - 752 MB | NPU
| recognizer | ONNX | float | Qualcomm® QCS8550 (Proxy) | 16.531 ms | 0 - 17 MB | NPU
| recognizer | ONNX | float | Qualcomm® QCS9075 | 19.253 ms | 0 - 3 MB | NPU
| recognizer | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 10.272 ms | 0 - 633 MB | NPU
| recognizer | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 9.266 ms | 1 - 874 MB | NPU
| recognizer | ONNX | w8a8 | Snapdragon® X2 Elite | 10.834 ms | 14 - 14 MB | NPU
| recognizer | ONNX | w8a8 | Snapdragon® X Elite | 12.333 ms | 10 - 10 MB | NPU
| recognizer | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 8.892 ms | 2 - 864 MB | NPU
| recognizer | ONNX | w8a8 | Qualcomm® QCS6490 | 120.647 ms | 21 - 24 MB | CPU
| recognizer | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 12.041 ms | 0 - 16 MB | NPU
| recognizer | ONNX | w8a8 | Qualcomm® QCS9075 | 14.271 ms | 2 - 5 MB | NPU
| recognizer | ONNX | w8a8 | Qualcomm® QCM6690 | 55.092 ms | 21 - 27 MB | CPU
| recognizer | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 8.305 ms | 2 - 704 MB | NPU
| recognizer | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 41.121 ms | 20 - 27 MB | CPU
| recognizer | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 10.453 ms | 0 - 722 MB | NPU
| recognizer | QNN_DLC | float | Snapdragon® X2 Elite | 12.503 ms | 0 - 0 MB | NPU
| recognizer | QNN_DLC | float | Snapdragon® X Elite | 16.347 ms | 0 - 0 MB | NPU
| recognizer | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 12.495 ms | 0 - 785 MB | NPU
| recognizer | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 47.981 ms | 0 - 669 MB | NPU
| recognizer | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 16.736 ms | 0 - 477 MB | NPU
| recognizer | QNN_DLC | float | Qualcomm® SA8775P | 21.111 ms | 0 - 672 MB | NPU
| recognizer | QNN_DLC | float | Qualcomm® QCS9075 | 18.559 ms | 2 - 4 MB | NPU
| recognizer | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 22.289 ms | 0 - 406 MB | NPU
| recognizer | QNN_DLC | float | Qualcomm® SA7255P | 47.981 ms | 0 - 669 MB | NPU
| recognizer | QNN_DLC | float | Qualcomm® SA8295P | 24.842 ms | 0 - 338 MB | NPU
| recognizer | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 10.379 ms | 0 - 678 MB | NPU
| recognizer | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 39.886 ms | 11 - 21 MB | CPU
| recognizer | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 51.863 ms | 7 - 16 MB | CPU
| recognizer | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 171.493 ms | 0 - 8 MB | CPU
| recognizer | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 65.319 ms | 6 - 7 MB | CPU
| recognizer | TFLITE | float | Qualcomm® SA8775P | 98.469 ms | 8 - 14 MB | CPU
| recognizer | TFLITE | float | Qualcomm® QCS9075 | 130.52 ms | 4 - 33 MB | CPU
| recognizer | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 70.962 ms | 6 - 17 MB | CPU
| recognizer | TFLITE | float | Qualcomm® SA7255P | 171.493 ms | 0 - 8 MB | CPU
| recognizer | TFLITE | float | Qualcomm® SA8295P | 71.334 ms | 8 - 13 MB | CPU
| recognizer | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 43.98 ms | 8 - 20 MB | CPU
| recognizer | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 27.193 ms | 9 - 19 MB | CPU
| recognizer | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 32.055 ms | 5 - 16 MB | CPU
| recognizer | TFLITE | w8a8 | Qualcomm® QCS6490 | 187.507 ms | 3 - 22 MB | CPU
| recognizer | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 142.899 ms | 8 - 16 MB | CPU
| recognizer | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 35.309 ms | 5 - 7 MB | CPU
| recognizer | TFLITE | w8a8 | Qualcomm® SA8775P | 70.046 ms | 7 - 13 MB | CPU
| recognizer | TFLITE | w8a8 | Qualcomm® QCS9075 | 100.451 ms | 3 - 21 MB | CPU
| recognizer | TFLITE | w8a8 | Qualcomm® QCM6690 | 74.848 ms | 6 - 12 MB | CPU
| recognizer | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 40.542 ms | 5 - 16 MB | CPU
| recognizer | TFLITE | w8a8 | Qualcomm® SA7255P | 142.899 ms | 8 - 16 MB | CPU
| recognizer | TFLITE | w8a8 | Qualcomm® SA8295P | 50.992 ms | 6 - 11 MB | CPU
| recognizer | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 28.775 ms | 7 - 16 MB | CPU
| recognizer | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 57.306 ms | 6 - 13 MB | CPU

## License
* The license for the original implementation of EasyOCR can be found
  [here](https://github.com/JaidedAI/EasyOCR/blob/master/LICENSE).

## References
* [Source Model Implementation](https://github.com/JaidedAI/EasyOCR)

## 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).