Hugging Face's logo

qualcomm
/
OpusMT-Zh-En

Text Generation
PyTorch
foundation
android
Model card Files
xet
Community

OpusMT-Zh-En: Optimized for Qualcomm Devices

OpusMT Chinese to English translation model is a state-of-the-art neural machine translation system designed for translating Chinese text into English. This model is based on the Marian transformer architecture and has been optimized for edge inference by splitting into encoder and decoder components with modified attention mechanisms. It exhibits robust performance for real-world translation tasks, making it highly reliable for practical applications. The model supports input sequences up to 256 tokens and can generate English translations with high accuracy.

This is based on the implementation of OpusMT-Zh-En 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 2.42, ONNX Runtime 1.24.1 Download
QNN_DLC float Universal QAIRT 2.43 Download
TFLITE float Universal QAIRT 2.43, TFLite 2.17.0 Download

For more device-specific assets and performance metrics, visit OpusMT-Zh-En 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 OpusMT-Zh-En on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.text_generation

Model Stats:

  • Model checkpoint: Helsinki-NLP/opus-mt-zh-en
  • Input resolution: 256 tokens (Chinese text)
  • Max input sequence length: 256 tokens
  • Max output sequence length: 256 tokens
  • Number of parameters (OpusMTEncoder): ~74M
  • Model size (OpusMTEncoder) (float): ~280 MB
  • Number of parameters (OpusMTDecoder): ~74M
  • Model size (OpusMTDecoder) (float): ~280 MB
  • Number of encoder layers: 6
  • Number of decoder layers: 6
  • Attention heads: 8
  • Hidden dimension: 512

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
OpusMTDecoder ONNX float Snapdragon® X Elite 3.311 ms 160 - 160 MB NPU
OpusMTDecoder ONNX float Snapdragon® 8 Gen 3 Mobile 2.752 ms 0 - 397 MB NPU
OpusMTDecoder ONNX float Qualcomm® QCS8550 (Proxy) 3.515 ms 12 - 17 MB NPU
OpusMTDecoder ONNX float Qualcomm® QCS9075 4.24 ms 12 - 27 MB NPU
OpusMTDecoder ONNX float Snapdragon® 8 Elite For Galaxy Mobile 2.358 ms 0 - 391 MB NPU
OpusMTDecoder ONNX float Snapdragon® 8 Elite Gen 5 Mobile 2.219 ms 1 - 378 MB NPU
OpusMTDecoder ONNX float Snapdragon® X2 Elite 1.807 ms 161 - 161 MB NPU
OpusMTDecoder QNN_DLC float Snapdragon® X Elite 3.082 ms 6 - 6 MB NPU
OpusMTDecoder QNN_DLC float Snapdragon® 8 Gen 3 Mobile 2.337 ms 0 - 189 MB NPU
OpusMTDecoder QNN_DLC float Qualcomm® QCS8275 (Proxy) 6.814 ms 2 - 167 MB NPU
OpusMTDecoder QNN_DLC float Qualcomm® QCS8550 (Proxy) 3.091 ms 0 - 2 MB NPU
OpusMTDecoder QNN_DLC float Qualcomm® SA8775P 3.945 ms 2 - 168 MB NPU
OpusMTDecoder QNN_DLC float Qualcomm® QCS9075 3.368 ms 6 - 14 MB NPU
OpusMTDecoder QNN_DLC float Qualcomm® QCS8450 (Proxy) 4.004 ms 6 - 173 MB NPU
OpusMTDecoder QNN_DLC float Qualcomm® SA7255P 6.814 ms 2 - 167 MB NPU
OpusMTDecoder QNN_DLC float Qualcomm® SA8295P 4.133 ms 6 - 155 MB NPU
OpusMTDecoder QNN_DLC float Snapdragon® 8 Elite For Galaxy Mobile 2.133 ms 0 - 165 MB NPU
OpusMTDecoder QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 1.869 ms 0 - 171 MB NPU
OpusMTDecoder QNN_DLC float Snapdragon® X2 Elite 1.858 ms 6 - 6 MB NPU
OpusMTDecoder TFLITE float Snapdragon® 8 Gen 3 Mobile 2.987 ms 0 - 381 MB NPU
OpusMTDecoder TFLITE float Qualcomm® QCS8275 (Proxy) 6.652 ms 0 - 229 MB NPU
OpusMTDecoder TFLITE float Qualcomm® QCS8550 (Proxy) 3.478 ms 0 - 3 MB NPU
OpusMTDecoder TFLITE float Qualcomm® SA8775P 4.234 ms 0 - 231 MB NPU
OpusMTDecoder TFLITE float Qualcomm® QCS9075 4.206 ms 0 - 186 MB NPU
OpusMTDecoder TFLITE float Qualcomm® QCS8450 (Proxy) 4.709 ms 0 - 354 MB NPU
OpusMTDecoder TFLITE float Qualcomm® SA7255P 6.652 ms 0 - 229 MB NPU
OpusMTDecoder TFLITE float Qualcomm® SA8295P 4.879 ms 0 - 212 MB NPU
OpusMTDecoder TFLITE float Snapdragon® 8 Elite For Galaxy Mobile 2.52 ms 0 - 414 MB NPU
OpusMTDecoder TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 2.148 ms 0 - 374 MB NPU
OpusMTEncoder ONNX float Snapdragon® X Elite 5.068 ms 108 - 108 MB NPU
OpusMTEncoder ONNX float Snapdragon® 8 Gen 3 Mobile 3.415 ms 0 - 261 MB NPU
OpusMTEncoder ONNX float Qualcomm® QCS8550 (Proxy) 4.761 ms 16 - 19 MB NPU
OpusMTEncoder ONNX float Qualcomm® QCS9075 6.208 ms 16 - 18 MB NPU
OpusMTEncoder ONNX float Snapdragon® 8 Elite For Galaxy Mobile 2.771 ms 0 - 290 MB NPU
OpusMTEncoder ONNX float Snapdragon® 8 Elite Gen 5 Mobile 2.023 ms 0 - 257 MB NPU
OpusMTEncoder ONNX float Snapdragon® X2 Elite 2.194 ms 108 - 108 MB NPU
OpusMTEncoder QNN_DLC float Snapdragon® X Elite 3.862 ms 0 - 0 MB NPU
OpusMTEncoder QNN_DLC float Snapdragon® 8 Gen 3 Mobile 2.543 ms 0 - 90 MB NPU
OpusMTEncoder QNN_DLC float Qualcomm® QCS8275 (Proxy) 12.6 ms 0 - 52 MB NPU
OpusMTEncoder QNN_DLC float Qualcomm® QCS8550 (Proxy) 3.536 ms 0 - 2 MB NPU
OpusMTEncoder QNN_DLC float Qualcomm® SA8775P 4.551 ms 0 - 54 MB NPU
OpusMTEncoder QNN_DLC float Qualcomm® QCS9075 4.68 ms 2 - 7 MB NPU
OpusMTEncoder QNN_DLC float Qualcomm® QCS8450 (Proxy) 4.759 ms 0 - 86 MB NPU
OpusMTEncoder QNN_DLC float Qualcomm® SA7255P 12.6 ms 0 - 52 MB NPU
OpusMTEncoder QNN_DLC float Qualcomm® SA8295P 5.363 ms 0 - 54 MB NPU
OpusMTEncoder QNN_DLC float Snapdragon® 8 Elite For Galaxy Mobile 2.141 ms 0 - 56 MB NPU
OpusMTEncoder QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 1.614 ms 0 - 58 MB NPU
OpusMTEncoder QNN_DLC float Snapdragon® X2 Elite 1.969 ms 0 - 0 MB NPU
OpusMTEncoder TFLITE float Snapdragon® 8 Gen 3 Mobile 2.658 ms 0 - 261 MB NPU
OpusMTEncoder TFLITE float Qualcomm® QCS8275 (Proxy) 12.782 ms 6 - 100 MB NPU
OpusMTEncoder TFLITE float Qualcomm® QCS8550 (Proxy) 3.688 ms 4 - 275 MB NPU
OpusMTEncoder TFLITE float Qualcomm® SA8775P 4.818 ms 0 - 93 MB NPU
OpusMTEncoder TFLITE float Qualcomm® QCS9075 5.138 ms 6 - 120 MB NPU
OpusMTEncoder TFLITE float Qualcomm® QCS8450 (Proxy) 5.047 ms 6 - 258 MB NPU
OpusMTEncoder TFLITE float Qualcomm® SA7255P 12.782 ms 6 - 100 MB NPU
OpusMTEncoder TFLITE float Qualcomm® SA8295P 5.638 ms 6 - 95 MB NPU
OpusMTEncoder TFLITE float Snapdragon® 8 Elite For Galaxy Mobile 2.229 ms 0 - 280 MB NPU
OpusMTEncoder TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 1.76 ms 0 - 243 MB NPU

License

  • The license for the original implementation of OpusMT-Zh-En can be found here.

References

Community

Downloads last month

-

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