Hugging Face's logo

qualcomm
/
OpusMT-Es-En

Text Generation
PyTorch
foundation
android
Model card Files
xet
Community
qaihm-bot commited on
Commit
5ae2d1d
·
verified ·
1 Parent(s): e814125

v0.45.0

Browse files

See https://github.com/quic/ai-hub-models/releases/v0.45.0 for changelog.

Files changed (10) hide show
  1. .gitattributes +2 -0
  2. LICENSE +1 -0
  3. OpusMT-Es-En_OpusMTDecoder_float.dlc +3 -0
  4. OpusMT-Es-En_OpusMTDecoder_float.onnx.zip +3 -0
  5. OpusMT-Es-En_OpusMTDecoder_float.tflite +3 -0
  6. OpusMT-Es-En_OpusMTEncoder_float.dlc +3 -0
  7. OpusMT-Es-En_OpusMTEncoder_float.onnx.zip +3 -0
  8. OpusMT-Es-En_OpusMTEncoder_float.tflite +3 -0
  9. README.md +274 -0
  10. tool-versions.yaml +4 -0
.gitattributes CHANGED
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ OpusMT-Es-En_OpusMTDecoder_float.dlc filter=lfs diff=lfs merge=lfs -text
37
+ OpusMT-Es-En_OpusMTEncoder_float.dlc filter=lfs diff=lfs merge=lfs -text
LICENSE ADDED
@@ -0,0 +1 @@
 
 
1
+ The license of the original trained model can be found at https://github.com/Helsinki-NLP/Opus-MT/blob/master/LICENSE.
OpusMT-Es-En_OpusMTDecoder_float.dlc ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7850d8a4b1543500689e12fecd2e009d4471d583e953b219611f674c38f6c443
3
+ size 178309004
OpusMT-Es-En_OpusMTDecoder_float.onnx.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:004baca6056fdd32810832fb1197168dc839905ab673f33afc3aa0097f25cbd4
3
+ size 328487989
OpusMT-Es-En_OpusMTDecoder_float.tflite ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:47cc4aba6ca339c3273d9858cef2a146ff11b7690bf6868eb7ce96dee5c7299c
3
+ size 223058404
OpusMT-Es-En_OpusMTEncoder_float.dlc ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e423cc470b71823bfd18976725eb8671fc8515a0885e53ddcd00af594b42cbee
3
+ size 111199780
OpusMT-Es-En_OpusMTEncoder_float.onnx.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2dc14928c41b0df7e175ff85ff608e8cfd41702d626936f77ea2f54a784ced91
3
+ size 206390890
OpusMT-Es-En_OpusMTEncoder_float.tflite ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fb53ce6f7a7c75db7c51da6c1c1d09327dca805736580a0bb68ec689f97825d6
3
+ size 222233484
README.md ADDED
@@ -0,0 +1,274 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: pytorch
3
+ license: other
4
+ tags:
5
+ - foundation
6
+ - android
7
+ pipeline_tag: text-generation
8
+
9
+ ---
10
+
11
+ ![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/opus_mt_es_en/web-assets/model_demo.png)
12
+
13
+ # OpusMT-Es-En: Optimized for Mobile Deployment
14
+ ## OpusMT Spanish to English neural machine translation model based on MarianMT transformer architecture
15
+
16
+
17
+ OpusMT Spanish to English translation model is a state-of-the-art neural machine translation system designed for translating Spanish 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.
18
+
19
+ This model is an implementation of OpusMT-Es-En found [here](https://github.com/Helsinki-NLP/Opus-MT).
20
+
21
+
22
+ This repository provides scripts to run OpusMT-Es-En on Qualcomm® devices.
23
+ More details on model performance across various devices, can be found
24
+ [here](https://aihub.qualcomm.com/models/opus_mt_es_en).
25
+
26
+
27
+
28
+ ### Model Details
29
+
30
+ - **Model Type:** Model_use_case.text_generation
31
+ - **Model Stats:**
32
+ - Model checkpoint: Helsinki-NLP/opus-mt-es-en
33
+ - Input resolution: 256 tokens (Spanish text)
34
+ - Max input sequence length: 256 tokens
35
+ - Max output sequence length: 256 tokens
36
+ - Number of parameters (OpusMTEncoder): ~74M
37
+ - Model size (OpusMTEncoder) (float): ~280 MB
38
+ - Number of parameters (OpusMTDecoder): ~74M
39
+ - Model size (OpusMTDecoder) (float): ~280 MB
40
+ - Number of encoder layers: 6
41
+ - Number of decoder layers: 6
42
+ - Attention heads: 8
43
+ - Hidden dimension: 512
44
+
45
+ | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
46
+ |---|---|---|---|---|---|---|---|---|
47
+ | OpusMTEncoder | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 12.795 ms | 6 - 181 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
48
+ | OpusMTEncoder | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 12.536 ms | 0 - 139 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
49
+ | OpusMTEncoder | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 5.075 ms | 0 - 335 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
50
+ | OpusMTEncoder | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 4.724 ms | 0 - 173 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
51
+ | OpusMTEncoder | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 3.6 ms | 0 - 2 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
52
+ | OpusMTEncoder | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 3.528 ms | 0 - 2 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
53
+ | OpusMTEncoder | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 5.721 ms | 0 - 154 MB | NPU | [OpusMT-Es-En.onnx.zip](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.onnx.zip) |
54
+ | OpusMTEncoder | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 19.66 ms | 6 - 182 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
55
+ | OpusMTEncoder | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 19.309 ms | 0 - 140 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
56
+ | OpusMTEncoder | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 12.795 ms | 6 - 181 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
57
+ | OpusMTEncoder | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 12.536 ms | 0 - 139 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
58
+ | OpusMTEncoder | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 5.545 ms | 6 - 180 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
59
+ | OpusMTEncoder | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 5.248 ms | 0 - 139 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
60
+ | OpusMTEncoder | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 19.66 ms | 6 - 182 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
61
+ | OpusMTEncoder | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 19.309 ms | 0 - 140 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
62
+ | OpusMTEncoder | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 2.656 ms | 0 - 336 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
63
+ | OpusMTEncoder | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 2.55 ms | 0 - 174 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
64
+ | OpusMTEncoder | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 4.135 ms | 15 - 329 MB | NPU | [OpusMT-Es-En.onnx.zip](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.onnx.zip) |
65
+ | OpusMTEncoder | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 2.241 ms | 0 - 356 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
66
+ | OpusMTEncoder | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 2.132 ms | 0 - 143 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
67
+ | OpusMTEncoder | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 3.54 ms | 0 - 335 MB | NPU | [OpusMT-Es-En.onnx.zip](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.onnx.zip) |
68
+ | OpusMTEncoder | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | TFLITE | 1.756 ms | 0 - 315 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
69
+ | OpusMTEncoder | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | QNN_DLC | 1.622 ms | 0 - 146 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
70
+ | OpusMTEncoder | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | ONNX | 2.846 ms | 0 - 299 MB | NPU | [OpusMT-Es-En.onnx.zip](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.onnx.zip) |
71
+ | OpusMTEncoder | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 3.838 ms | 0 - 0 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
72
+ | OpusMTEncoder | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 5.837 ms | 109 - 109 MB | NPU | [OpusMT-Es-En.onnx.zip](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.onnx.zip) |
73
+ | OpusMTDecoder | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 6.446 ms | 0 - 424 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
74
+ | OpusMTDecoder | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 5.416 ms | 6 - 241 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
75
+ | OpusMTDecoder | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 4.795 ms | 0 - 426 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
76
+ | OpusMTDecoder | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 3.987 ms | 2 - 247 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
77
+ | OpusMTDecoder | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 3.469 ms | 0 - 3 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
78
+ | OpusMTDecoder | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 2.891 ms | 2 - 3 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
79
+ | OpusMTDecoder | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 3.802 ms | 12 - 14 MB | NPU | [OpusMT-Es-En.onnx.zip](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.onnx.zip) |
80
+ | OpusMTDecoder | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 4.132 ms | 0 - 292 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
81
+ | OpusMTDecoder | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 3.495 ms | 6 - 243 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
82
+ | OpusMTDecoder | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 6.446 ms | 0 - 424 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
83
+ | OpusMTDecoder | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 5.416 ms | 6 - 241 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
84
+ | OpusMTDecoder | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 4.792 ms | 0 - 284 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
85
+ | OpusMTDecoder | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 4.042 ms | 6 - 234 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
86
+ | OpusMTDecoder | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 4.132 ms | 0 - 292 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
87
+ | OpusMTDecoder | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 3.495 ms | 6 - 243 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
88
+ | OpusMTDecoder | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 2.649 ms | 0 - 573 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
89
+ | OpusMTDecoder | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 2.199 ms | 0 - 258 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
90
+ | OpusMTDecoder | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 2.923 ms | 0 - 417 MB | NPU | [OpusMT-Es-En.onnx.zip](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.onnx.zip) |
91
+ | OpusMTDecoder | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 2.303 ms | 0 - 489 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
92
+ | OpusMTDecoder | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 1.955 ms | 0 - 236 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
93
+ | OpusMTDecoder | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 2.59 ms | 0 - 441 MB | NPU | [OpusMT-Es-En.onnx.zip](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.onnx.zip) |
94
+ | OpusMTDecoder | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | TFLITE | 2.165 ms | 0 - 433 MB | NPU | [OpusMT-Es-En.tflite](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.tflite) |
95
+ | OpusMTDecoder | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | QNN_DLC | 1.896 ms | 0 - 243 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
96
+ | OpusMTDecoder | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen 5 Mobile | ONNX | 2.506 ms | 1 - 406 MB | NPU | [OpusMT-Es-En.onnx.zip](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.onnx.zip) |
97
+ | OpusMTDecoder | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 2.745 ms | 6 - 6 MB | NPU | [OpusMT-Es-En.dlc](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.dlc) |
98
+ | OpusMTDecoder | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 3.15 ms | 161 - 161 MB | NPU | [OpusMT-Es-En.onnx.zip](https://huggingface.co/qualcomm/OpusMT-Es-En/blob/main/OpusMT-Es-En.onnx.zip) |
99
+
100
+
101
+
102
+
103
+ ## Installation
104
+
105
+
106
+ Install the package via pip:
107
+ ```bash
108
+ # NOTE: 3.10 <= PYTHON_VERSION < 3.14 is supported.
109
+ pip install "qai-hub-models[opus-mt-es-en]"
110
+ ```
111
+
112
+
113
+ ## Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device
114
+
115
+ Sign-in to [Qualcomm® AI Hub Workbench](https://workbench.aihub.qualcomm.com/) with your
116
+ Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`.
117
+
118
+ With this API token, you can configure your client to run models on the cloud
119
+ hosted devices.
120
+ ```bash
121
+ qai-hub configure --api_token API_TOKEN
122
+ ```
123
+ Navigate to [docs](https://workbench.aihub.qualcomm.com/docs/) for more information.
124
+
125
+
126
+
127
+ ## Demo off target
128
+
129
+ The package contains a simple end-to-end demo that downloads pre-trained
130
+ weights and runs this model on a sample input.
131
+
132
+ ```bash
133
+ python -m qai_hub_models.models.opus_mt_es_en.demo
134
+ ```
135
+
136
+ The above demo runs a reference implementation of pre-processing, model
137
+ inference, and post processing.
138
+
139
+ **NOTE**: If you want running in a Jupyter Notebook or Google Colab like
140
+ environment, please add the following to your cell (instead of the above).
141
+ ```
142
+ %run -m qai_hub_models.models.opus_mt_es_en.demo
143
+ ```
144
+
145
+
146
+ ### Run model on a cloud-hosted device
147
+
148
+ In addition to the demo, you can also run the model on a cloud-hosted Qualcomm®
149
+ device. This script does the following:
150
+ * Performance check on-device on a cloud-hosted device
151
+ * Downloads compiled assets that can be deployed on-device for Android.
152
+ * Accuracy check between PyTorch and on-device outputs.
153
+
154
+ ```bash
155
+ python -m qai_hub_models.models.opus_mt_es_en.export
156
+ ```
157
+
158
+
159
+
160
+ ## How does this work?
161
+
162
+ This [export script](https://aihub.qualcomm.com/models/opus_mt_es_en/qai_hub_models/models/OpusMT-Es-En/export.py)
163
+ leverages [Qualcomm® AI Hub](https://aihub.qualcomm.com/) to optimize, validate, and deploy this model
164
+ on-device. Lets go through each step below in detail:
165
+
166
+ Step 1: **Compile model for on-device deployment**
167
+
168
+ To compile a PyTorch model for on-device deployment, we first trace the model
169
+ in memory using the `jit.trace` and then call the `submit_compile_job` API.
170
+
171
+ ```python
172
+ import torch
173
+
174
+ import qai_hub as hub
175
+ from qai_hub_models.models.opus_mt_es_en import Model
176
+
177
+ # Load the model
178
+ torch_model = Model.from_pretrained()
179
+
180
+ # Device
181
+ device = hub.Device("Samsung Galaxy S25")
182
+
183
+ # Trace model
184
+ input_shape = torch_model.get_input_spec()
185
+ sample_inputs = torch_model.sample_inputs()
186
+
187
+ pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])
188
+
189
+ # Compile model on a specific device
190
+ compile_job = hub.submit_compile_job(
191
+ model=pt_model,
192
+ device=device,
193
+ input_specs=torch_model.get_input_spec(),
194
+ )
195
+
196
+ # Get target model to run on-device
197
+ target_model = compile_job.get_target_model()
198
+
199
+ ```
200
+
201
+
202
+ Step 2: **Performance profiling on cloud-hosted device**
203
+
204
+ After compiling models from step 1. Models can be profiled model on-device using the
205
+ `target_model`. Note that this scripts runs the model on a device automatically
206
+ provisioned in the cloud. Once the job is submitted, you can navigate to a
207
+ provided job URL to view a variety of on-device performance metrics.
208
+ ```python
209
+ profile_job = hub.submit_profile_job(
210
+ model=target_model,
211
+ device=device,
212
+ )
213
+
214
+ ```
215
+
216
+ Step 3: **Verify on-device accuracy**
217
+
218
+ To verify the accuracy of the model on-device, you can run on-device inference
219
+ on sample input data on the same cloud hosted device.
220
+ ```python
221
+ input_data = torch_model.sample_inputs()
222
+ inference_job = hub.submit_inference_job(
223
+ model=target_model,
224
+ device=device,
225
+ inputs=input_data,
226
+ )
227
+ on_device_output = inference_job.download_output_data()
228
+
229
+ ```
230
+ With the output of the model, you can compute like PSNR, relative errors or
231
+ spot check the output with expected output.
232
+
233
+ **Note**: This on-device profiling and inference requires access to Qualcomm®
234
+ AI Hub Workbench. [Sign up for access](https://myaccount.qualcomm.com/signup).
235
+
236
+
237
+
238
+
239
+ ## Deploying compiled model to Android
240
+
241
+
242
+ The models can be deployed using multiple runtimes:
243
+ - TensorFlow Lite (`.tflite` export): [This
244
+ tutorial](https://www.tensorflow.org/lite/android/quickstart) provides a
245
+ guide to deploy the .tflite model in an Android application.
246
+
247
+
248
+ - QNN (`.so` export ): This [sample
249
+ app](https://docs.qualcomm.com/bundle/publicresource/topics/80-63442-50/sample_app.html)
250
+ provides instructions on how to use the `.so` shared library in an Android application.
251
+
252
+
253
+ ## View on Qualcomm® AI Hub
254
+ Get more details on OpusMT-Es-En's performance across various devices [here](https://aihub.qualcomm.com/models/opus_mt_es_en).
255
+ Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
256
+
257
+
258
+ ## License
259
+ * The license for the original implementation of OpusMT-Es-En can be found
260
+ [here](https://github.com/Helsinki-NLP/Opus-MT/blob/master/LICENSE).
261
+
262
+
263
+
264
+ ## References
265
+ * [OPUS-MT – Building open translation services for the World](https://aclanthology.org/2020.eamt-1.61/)
266
+ * [Source Model Implementation](https://github.com/Helsinki-NLP/Opus-MT)
267
+
268
+
269
+
270
+ ## Community
271
+ * Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
272
+ * For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).
273
+
274
+
tool-versions.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ tool_versions:
2
+ onnx:
3
+ qairt: 2.37.1.250807093845_124904
4
+ onnx_runtime: 1.23.0