update doc (#3424)

docs/module_usage/tutorials/cv_modules/human_keypoint_detection.en.md

@@ -17,7 +17,7 @@ Keypoint detection algorithms mainly include two approaches: Top-Down and Bottom
     <th>Approach</th>
     <th>Input Size</th>
     <th>AP(0.5:0.95)</th>
-    <th>GPU Inference Time (ms)</th>
+    <th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
     <th>CPU Inference Time (ms)</th>
     <th>Model Size (M)</th>
     <th>Introduction</th>

docs/module_usage/tutorials/cv_modules/open_vocabulary_detection.en.md

@@ -14,7 +14,7 @@ Open-vocabulary object detection is an advanced object detection technology aime
 <th>Model</th><th>Model Download Link</th>
 <th>mAP(0.5:0.95)</th>
 <th>mAP(0.5)</th>
-<th>GPU Inference Time (ms)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU Inference Time (ms)</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>

docs/module_usage/tutorials/cv_modules/open_vocabulary_segmentation.en.md

@@ -12,7 +12,7 @@ Open-vocabulary segmentation is an image segmentation task that aims to segment
 <table>
 <tr>
 <th>Model</th><th>Model Download Link</th>
-<th>GPU Inference Time (ms)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU Inference Time (ms)</th>
 <th>Model Size (M)</th>
 <th>Description</th>

docs/module_usage/tutorials/cv_modules/rotated_object_detection.en.md

@@ -13,7 +13,7 @@ Rotated object detection is a derivative of the object detection module, specifi
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>mAP(%)</th>
-<th>GPU Inference Time (ms)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU Inference Time (ms)</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>

docs/module_usage/tutorials/cv_modules/semantic_segmentation.en.md

@@ -156,8 +156,8 @@ Semantic segmentation is a technique in computer vision that classifies each pix
 <tr>
 <th>Model Name</th><th>Model Download Link</th>
 <th>mIoU (%)</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Size (M)</th>
 </tr>
 </thead>

docs/module_usage/tutorials/ocr_modules/text_recognition.en.md

@@ -21,7 +21,7 @@ The text recognition module is the core component of an OCR (Optical Character R
 <tr>
 <td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_doc_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>81.53</td>
-<td>6.65 / 6.65</td>
+<td>6.65 / 2.38</td>
 <td>32.92 / 32.92</td>
 <td>74.7 M</td>
 <td>PP-OCRv4_server_rec_doc is trained on a mixed dataset of more Chinese document data and PP-OCR training data based on PP-OCRv4_server_rec. It has added the ability to recognize some traditional Chinese characters, Japanese, and special characters, and can support the recognition of more than 15,000 characters. In addition to improving the text recognition capability related to documents, it also enhances the general text recognition capability.</td>
@@ -29,7 +29,7 @@ The text recognition module is the core component of an OCR (Optical Character R
 <tr>
 <td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Training Model</a></td>
 <td>78.74</td>
-<td>4.82 / 4.82</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
 <td>
@@ -38,7 +38,7 @@ The lightweight recognition model of PP-OCRv4 has high inference efficiency and
 <tr>
 <td>PP-OCRv4_server_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Training Model</a></td>
 <td>80.61 </td>
-<td>6.58 / 6.58</td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
 <td>The server-side model of PP-OCRv4 offers high inference accuracy and can be deployed on various types of servers.</td>
@@ -46,7 +46,7 @@ The lightweight recognition model of PP-OCRv4 has high inference efficiency and
 <tr>
 <td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>70.39</td>
-<td>4.81 / 4.81</td>
+<td>4.81 / 0.75</td>
 <td>16.10 / 5.31</td>
 <td>6.8 M</td>
 <td>The ultra-lightweight English recognition model, trained based on the PP-OCRv4 recognition model, supports the recognition of English letters and numbers.</td>
@@ -70,7 +70,7 @@ The lightweight recognition model of PP-OCRv4 has high inference efficiency and
 <tr>
 <td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_doc_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>81.53</td>
-<td>6.65 / 6.65</td>
+<td>6.65 / 2.38</td>
 <td>32.92 / 32.92</td>
 <td>74.7 M</td>
 <td>PP-OCRv4_server_rec_doc is trained on a mixed dataset of more Chinese document data and PP-OCR training data based on PP-OCRv4_server_rec. It has added the recognition capabilities for some traditional Chinese characters, Japanese, and special characters. The number of recognizable characters is over 15,000. In addition to the improvement in document-related text recognition, it also enhances the general text recognition capability.</td>
@@ -78,7 +78,7 @@ The lightweight recognition model of PP-OCRv4 has high inference efficiency and
 <tr>
 <td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Training Model</a></td>
 <td>78.74</td>
-<td>4.82 / 4.82</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
 <td>The lightweight recognition model of PP-OCRv4 has high inference efficiency and can be deployed on various hardware devices, including edge devices.</td>
@@ -86,7 +86,7 @@ The lightweight recognition model of PP-OCRv4 has high inference efficiency and
 <tr>
 <td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Trained Model</a></td>
 <td>80.61 </td>
-<td>6.58 / 6.58</td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
 <td>The server-side model of PP-OCRv4 offers high inference accuracy and can be deployed on various types of servers.</td>
@@ -94,7 +94,7 @@ The lightweight recognition model of PP-OCRv4 has high inference efficiency and
 <tr>
 <td>PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>72.96</td>
-<td>5.87 / 5.87</td>
+<td>5.87 / 1.19</td>
 <td>9.07 / 4.28</td>
 <td>9.2 M</td>
 <td>PP-OCRv3’s lightweight recognition model is designed for high inference efficiency and can be deployed on a variety of hardware devices, including edge devices.</td>
@@ -105,15 +105,15 @@ The lightweight recognition model of PP-OCRv4 has high inference efficiency and
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Recognition Avg Accuracy(%)</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <tr>
 <td>ch_SVTRv2_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_SVTRv2_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_SVTRv2_rec_pretrained.pdparams">Training Model</a></td>
 <td>68.81</td>
-<td>8.08 / 8.08</td>
+<td>8.08 / 2.74</td>
 <td>50.17 / 42.50</td>
 <td>73.9 M</td>
 <td rowspan="1">
@@ -126,15 +126,15 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Recognition Avg Accuracy(%)</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <tr>
 <td>ch_RepSVTR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_RepSVTR_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_RepSVTR_rec_pretrained.pdparams">Training Model</a></td>
 <td>65.07</td>
-<td>5.93 / 5.93</td>
+<td>5.93 / 1.62</td>
 <td>20.73 / 7.32</td>
 <td>22.1 M</td>
 <td rowspan="1">    The RepSVTR text recognition model is a mobile text recognition model based on SVTRv2. It won the first prize in the PaddleOCR Algorithm Model Challenge - Task One: OCR End-to-End Recognition Task. The end-to-end recognition accuracy on the B list is 2.5% higher than that of PP-OCRv4, with the same inference speed.</td>
@@ -146,15 +146,15 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Recognition Avg Accuracy(%)</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <tr>
 <td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td> 70.39</td>
-<td>4.81 / 4.81</td>
+<td>4.81 / 0.75</td>
 <td>16.10 / 5.31</td>
 <td>6.8 M</td>
 <td>The ultra-lightweight English recognition model trained based on the PP-OCRv4 recognition model supports the recognition of English and numbers.</td>
@@ -162,7 +162,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>en_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>70.69</td>
-<td>5.44 / 5.44</td>
+<td>5.44 / 0.75</td>
 <td>8.65 / 5.57</td>
 <td>7.8 M </td>
 <td>The ultra-lightweight English recognition model trained based on the PP-OCRv3 recognition model supports the recognition of English and numbers.</td>
@@ -174,15 +174,15 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Recognition Avg Accuracy(%)</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <tr>
 <td>korean_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/korean_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>60.21</td>
-<td>5.40 / 5.40</td>
+<td>5.40 / 0.97</td>
 <td>9.11 / 4.05</td>
 <td>8.6 M</td>
 <td>The ultra-lightweight Korean recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Korean and numbers. </td>
@@ -190,7 +190,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>japan_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/japan_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>45.69</td>
-<td>5.70 / 5.70</td>
+<td>5.70 / 1.02</td>
 <td>8.48 / 4.07</td>
 <td>8.8 M </td>
 <td>The ultra-lightweight Japanese recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Japanese and numbers.</td>
@@ -198,7 +198,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>chinese_cht_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/chinese_cht_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>82.06</td>
-<td>5.90 / 5.90</td>
+<td>5.90 / 1.28</td>
 <td>9.28 / 4.34</td>
 <td>9.7 M </td>
 <td>The ultra-lightweight Traditional Chinese recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Traditional Chinese and numbers.</td>
@@ -206,7 +206,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>te_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/te_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>95.88</td>
-<td>5.42 / 5.42</td>
+<td>5.42 / 0.82</td>
 <td>8.10 / 6.91</td>
 <td>7.8 M </td>
 <td>The ultra-lightweight Telugu recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Telugu and numbers.</td>
@@ -214,7 +214,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>ka_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ka_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>96.96</td>
-<td>5.25 / 5.25</td>
+<td>5.25 / 0.79</td>
 <td>9.09 / 3.86</td>
 <td>8.0 M </td>
 <td>The ultra-lightweight Kannada recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Kannada and numbers.</td>
@@ -222,7 +222,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>ta_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ta_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>76.83</td>
-<td>5.23 / 5.23</td>
+<td>5.23 / 0.75</td>
 <td>10.13 / 4.30</td>
 <td>8.0 M </td>
 <td>The ultra-lightweight Tamil recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Tamil and numbers.</td>
@@ -230,7 +230,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>latin_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/latin_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>76.93</td>
-<td>5.20 / 5.20</td>
+<td>5.20 / 0.79</td>
 <td>8.83 / 7.15</td>
 <td>7.8 M</td>
 <td>The ultra-lightweight Latin recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Latin script and numbers.</td>
@@ -238,7 +238,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>arabic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/arabic_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>73.55</td>
-<td>5.35 / 5.35</td>
+<td>5.35 / 0.79</td>
 <td>8.80 / 4.56</td>
 <td>7.8 M</td>
 <td>The ultra-lightweight Arabic script recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Arabic script and numbers.</td>
@@ -246,7 +246,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>cyrillic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/cyrillic_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>94.28</td>
-<td>5.23 / 5.23</td>
+<td>5.23 / 0.76</td>
 <td>8.89 / 3.88</td>
 <td>7.9 M  </td>
 <td>
@@ -255,7 +255,7 @@ The ultra-lightweight cyrillic alphabet recognition model trained based on the P
 <tr>
 <td>devanagari_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/devanagari_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>96.44</td>
-<td>5.22 / 5.22</td>
+<td>5.22 / 0.79</td>
 <td>8.56 / 4.06</td>
 <td>7.9 M  </td>
 <td>The ultra-lightweight Devanagari script recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Devanagari script and numbers.</td>

docs/module_usage/tutorials/ocr_modules/text_recognition.md

@@ -22,7 +22,7 @@ comments: true
 <td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
 PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_doc_pretrained.pdparams">训练模型</a></td>
 <td>81.53</td>
-<td>6.65 / 6.65</td>
+<td>6.65 / 2.38</td>
 <td>32.92 / 32.92</td>
 <td>74.7 M</td>
 <td>PP-OCRv4_server_rec_doc是在PP-OCRv4_server_rec的基础上，在更多中文文档数据和PP-OCR训练数据的混合数据训练而成，增加了部分繁体字、日文、特殊字符的识别能力，可支持识别的字符为1.5万+，除文档相关的文字识别能力提升外，也同时提升了通用文字的识别能力</td>
@@ -30,7 +30,7 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="https://paddle-mode
 <tr>
 <td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>78.74</td>
-<td>4.82 / 4.82</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
 <td>PP-OCRv4的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
@@ -38,7 +38,7 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="https://paddle-mode
 <tr>
 <td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">训练模型</a></td>
 <td>80.61 </td>
-<td>6.58 / 6.58</td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
 <td>PP-OCRv4的服务器端模型，推理精度高，可以部署在多种不同的服务器上</td>
@@ -47,7 +47,7 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="https://paddle-mode
 <td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
 en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>70.39</td>
-<td>4.81 / 4.81</td>
+<td>4.81 / 0.75</td>
 <td>16.10 / 5.31</td>
 <td>6.8 M</td>
 <td>基于PP-OCRv4识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
@@ -72,7 +72,7 @@ en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model
 <td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
 PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_doc_pretrained.pdparams">训练模型</a></td>
 <td>81.53</td>
-<td>6.65 / 6.65</td>
+<td>6.65 / 2.38</td>
 <td>32.92 / 32.92</td>
 <td>74.7 M</td>
 <td>PP-OCRv4_server_rec_doc是在PP-OCRv4_server_rec的基础上，在更多中文文档数据和PP-OCR训练数据的混合数据训练而成，增加了部分繁体字、日文、特殊字符的识别能力，可支持识别的字符为1.5万+，除文档相关的文字识别能力提升外，也同时提升了通用文字的识别能力</td>
@@ -80,7 +80,7 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="https://paddle-mode
 <tr>
 <td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>78.74</td>
-<td>4.82 / 4.82</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
 <td>PP-OCRv4的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
@@ -88,7 +88,7 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="https://paddle-mode
 <tr>
 <td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">训练模型</a></td>
 <td>80.61 </td>
-<td>6.58 / 6.58</td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
 <td>PP-OCRv4的服务器端模型，推理精度高，可以部署在多种不同的服务器上</td>
@@ -97,7 +97,7 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="https://paddle-mode
 <td>PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
 PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>72.96</td>
-<td>5.87 / 5.87</td>
+<td>5.87 / 1.19</td>
 <td>9.07 / 4.28</td>
 <td>9.2 M</td>
 <td>PP-OCRv3的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
@@ -116,7 +116,7 @@ PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ec
 <tr>
 <td>ch_SVTRv2_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_SVTRv2_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_SVTRv2_rec_pretrained.pdparams">训练模型</a></td>
 <td>68.81</td>
-<td>8.08 / 8.08</td>
+<td>8.08 / 2.74</td>
 <td>50.17 / 42.50</td>
 <td>73.9 M</td>
 <td rowspan="1">
@@ -137,7 +137,7 @@ SVTRv2 是一种由复旦大学视觉与学习实验室（FVL）的OpenOCR团队
 <tr>
 <td>ch_RepSVTR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_RepSVTR_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_RepSVTR_rec_pretrained.pdparams">训练模型</a></td>
 <td>65.07</td>
-<td>5.93 / 5.93</td>
+<td>5.93 / 1.62</td>
 <td>20.73 / 7.32</td>
 <td>22.1 M</td>
 <td rowspan="1">    RepSVTR 文本识别模型是一种基于SVTRv2 的移动端文本识别模型，其在PaddleOCR算法模型挑战赛 - 赛题一：OCR端到端识别任务中荣获一等奖，B榜端到端识别精度相比PP-OCRv4提升2.5%，推理速度持平。</td>
@@ -158,7 +158,7 @@ SVTRv2 是一种由复旦大学视觉与学习实验室（FVL）的OpenOCR团队
 <td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
 en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td> 70.39</td>
-<td>4.81 / 4.81</td>
+<td>4.81 / 0.75</td>
 <td>16.10 / 5.31</td>
 <td>6.8 M</td>
 <td>基于PP-OCRv4识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
@@ -167,7 +167,7 @@ en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model
 <td>en_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
 en_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>70.69</td>
-<td>5.44 / 5.44</td>
+<td>5.44 / 0.75</td>
 <td>8.65 / 5.57</td>
 <td>7.8 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
@@ -189,7 +189,7 @@ en_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model
 <td>korean_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
 korean_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/korean_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>60.21</td>
-<td>5.40 / 5.40</td>
+<td>5.40 / 0.97</td>
 <td>9.11 / 4.05</td>
 <td>8.6 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量韩文识别模型，支持韩文、数字识别</td>
@@ -198,7 +198,7 @@ korean_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-m
 <td>japan_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
 japan_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/japan_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>45.69</td>
-<td>5.70 / 5.70</td>
+<td>5.70 / 1.02</td>
 <td>8.48 / 4.07</td>
 <td>8.8 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量日文识别模型，支持日文、数字识别</td>
@@ -207,7 +207,7 @@ japan_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-mo
 <td>chinese_cht_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
 chinese_cht_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/chinese_cht_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>82.06</td>
-<td>5.90 / 5.90</td>
+<td>5.90 / 1.28</td>
 <td>9.28 / 4.34</td>
 <td>9.7 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量繁体中文识别模型，支持繁体中文、数字识别</td>
@@ -216,7 +216,7 @@ chinese_cht_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://pad
 <td>te_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
 te_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/te_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>95.88</td>
-<td>5.42 / 5.42</td>
+<td>5.42 / 0.82</td>
 <td>8.10 / 6.91</td>
 <td>7.8 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量泰卢固文识别模型，支持泰卢固文、数字识别</td>
@@ -225,7 +225,7 @@ te_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model
 <td>ka_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
 ka_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ka_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>96.96</td>
-<td>5.25 / 5.25</td>
+<td>5.25 / 0.79</td>
 <td>9.09 / 3.86</td>
 <td>8.0 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量卡纳达文识别模型，支持卡纳达文、数字识别</td>
@@ -234,7 +234,7 @@ ka_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model
 <td>ta_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
 ta_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ta_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>76.83</td>
-<td>5.23 / 5.23</td>
+<td>5.23 / 0.75</td>
 <td>10.13 / 4.30</td>
 <td>8.0 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量泰米尔文识别模型，支持泰米尔文、数字识别</td>
@@ -243,7 +243,7 @@ ta_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model
 <td>latin_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
 latin_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/latin_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>76.93</td>
-<td>5.20 / 5.20</td>
+<td>5.20 / 0.79</td>
 <td>8.83 / 7.15</td>
 <td>7.8 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量拉丁文识别模型，支持拉丁文、数字识别</td>
@@ -252,7 +252,7 @@ latin_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-mo
 <td>arabic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
 arabic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/arabic_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>73.55</td>
-<td>5.35 / 5.35</td>
+<td>5.35 / 0.79</td>
 <td>8.80 / 4.56</td>
 <td>7.8 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量阿拉伯字母识别模型，支持阿拉伯字母、数字识别</td>
@@ -261,7 +261,7 @@ arabic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-m
 <td>cyrillic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
 cyrillic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/cyrillic_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>94.28</td>
-<td>5.23 / 5.23</td>
+<td>5.23 / 0.76</td>
 <td>8.89 / 3.88</td>
 <td>7.9 M  </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量斯拉夫字母识别模型，支持斯拉夫字母、数字识别</td>
@@ -270,7 +270,7 @@ cyrillic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle
 <td>devanagari_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
 devanagari_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/devanagari_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>96.44</td>
-<td>5.22 / 5.22</td>
+<td>5.22 / 0.79</td>
 <td>8.56 / 4.06</td>
 <td>7.9 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量梵文字母识别模型，支持梵文字母、数字识别</td>

docs/module_usage/tutorials/ocr_modules/textline_orientation_classification.en.md

@@ -14,7 +14,7 @@ The text line orientation classification module primarily distinguishes the orie
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Top-1 Accuracy (%)</th>
-<th>GPU Inference Time (ms)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU Inference Time (ms)</th>
 <th>Model Size (M)</th>
 <th>Description</th>

docs/pipeline_usage/tutorials/cv_pipelines/face_recognition.en.md

@@ -68,8 +68,8 @@ The face recognition pipeline is an end-to-end system dedicated to solving face
 <th>Model</th><th>Model Download Link</th>
 <th>Output Feature Dimension</th>
 <th>Acc (%)<br/>AgeDB-30/CFP-FP/LFW</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Size (M)</th>
 <th>Description</th>
 </tr>

docs/pipeline_usage/tutorials/cv_pipelines/open_vocabulary_detection.en.md

@@ -20,7 +20,7 @@ Open vocabulary object detection is an advanced object detection technology that
 <th>Model</th><th>Model Download Link</th>
 <th>mAP(0.5:0.95)</th>
 <th>mAP(0.5)</th>
-<th>GPU Inference Time (ms)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU Inference Time (ms)</th>
 <th>Model Storage Size (M)</th>
 <th>Description</th>

docs/pipeline_usage/tutorials/cv_pipelines/open_vocabulary_segmentation.en.md

@@ -18,7 +18,7 @@ Open vocabulary segmentation is an image segmentation task that aims to segment
 <table>
 <tr>
 <th>Model</th><th>Model Download Link</th>
-<th>GPU Inference Time (ms)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU Inference Time (ms)</th>
 <th>Model Storage Size (M)</th>
 <th>Description</th>

docs/pipeline_usage/tutorials/cv_pipelines/rotated_object_detection.en.md

@@ -19,7 +19,7 @@ Rotated object detection is a variant of the object detection module, specifical
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>mAP(%)</th>
-<th>GPU Inference Time (ms)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU Inference Time (ms)</th>
 <th>Model Storage Size (M)</th>
 <th>Description</th>

docs/pipeline_usage/tutorials/cv_pipelines/semantic_segmentation.en.md

@@ -156,8 +156,8 @@ Semantic segmentation is a computer vision technique that aims to assign each pi
 <tr>
 <th>Model Name</th><th>Model Download Link</th>
 <th>mIoU (%)</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Size (M)</th>
 </tr>
 </thead>

docs/pipeline_usage/tutorials/ocr_pipelines/OCR.en.md

@@ -109,7 +109,7 @@ The General OCR pipeline is designed to solve text recognition tasks, extracting
 <table>
 <tr>
 <th>Model</th><th>Model Download Link</th>
-<th>Recognition Avg Accuracy (%)</th>
+<th>Recognition Avg Accuracy(%)</th>
 <th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
@@ -118,34 +118,35 @@ The General OCR pipeline is designed to solve text recognition tasks, extracting
 <tr>
 <td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_doc_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>81.53</td>
-<td>6.65 / 6.65</td>
+<td>6.65 / 2.38</td>
 <td>32.92 / 32.92</td>
 <td>74.7 M</td>
-<td>PP-OCRv4_server_rec_doc is trained on a mixed dataset of more Chinese document data and PP-OCR training data, based on PP-OCRv4_server_rec. It enhances the recognition of traditional Chinese characters, Japanese, and special characters, supporting over 15,000 characters. It improves both document-related and general text recognition capabilities.</td>
+<td>PP-OCRv4_server_rec_doc is trained on a mixed dataset of more Chinese document data and PP-OCR training data based on PP-OCRv4_server_rec. It has added the ability to recognize some traditional Chinese characters, Japanese, and special characters, and can support the recognition of more than 15,000 characters. In addition to improving the text recognition capability related to documents, it also enhances the general text recognition capability.</td>
 </tr>
 <tr>
 <td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Training Model</a></td>
 <td>78.74</td>
-<td>4.82 / 4.82</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
-<td>The lightweight recognition model of PP-OCRv4, with high inference efficiency, suitable for deployment on various hardware devices, including edge devices</td>
+<td>
+The lightweight recognition model of PP-OCRv4 has high inference efficiency and can be deployed on various hardware devices, including edge devices.</td>
 </tr>
 <tr>
 <td>PP-OCRv4_server_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Training Model</a></td>
-<td>80.61</td>
-<td>6.58 / 6.58</td>
+<td>80.61 </td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
-<td>The server-side recognition model of PP-OCRv4, with high inference accuracy, suitable for deployment on various servers</td>
+<td>The server-side model of PP-OCRv4 offers high inference accuracy and can be deployed on various types of servers.</td>
 </tr>
 <tr>
 <td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>70.39</td>
-<td>4.81 / 4.81</td>
+<td>4.81 / 0.75</td>
 <td>16.10 / 5.31</td>
 <td>6.8 M</td>
-<td>The ultra-lightweight English recognition model trained based on the PP-OCRv4 recognition model, supporting English and numeric recognition</td>
+<td>The ultra-lightweight English recognition model, trained based on the PP-OCRv4 recognition model, supports the recognition of English letters and numbers.</td>
 </tr>
 </table>
 
@@ -166,7 +167,7 @@ The General OCR pipeline is designed to solve text recognition tasks, extracting
 <tr>
 <td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_doc_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>81.53</td>
-<td>6.65 / 6.65</td>
+<td>6.65 / 2.38</td>
 <td>32.92 / 32.92</td>
 <td>74.7 M</td>
 <td>PP-OCRv4_server_rec_doc is trained on a mixed dataset of more Chinese document data and PP-OCR training data based on PP-OCRv4_server_rec. It has added the recognition capabilities for some traditional Chinese characters, Japanese, and special characters. The number of recognizable characters is over 15,000. In addition to the improvement in document-related text recognition, it also enhances the general text recognition capability.</td>
@@ -174,7 +175,7 @@ The General OCR pipeline is designed to solve text recognition tasks, extracting
 <tr>
 <td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Training Model</a></td>
 <td>78.74</td>
-<td>4.82 / 4.82</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
 <td>The lightweight recognition model of PP-OCRv4 has high inference efficiency and can be deployed on various hardware devices, including edge devices.</td>
@@ -182,7 +183,7 @@ The General OCR pipeline is designed to solve text recognition tasks, extracting
 <tr>
 <td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Trained Model</a></td>
 <td>80.61 </td>
-<td>6.58 / 6.58</td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
 <td>The server-side model of PP-OCRv4 offers high inference accuracy and can be deployed on various types of servers.</td>
@@ -190,7 +191,7 @@ The General OCR pipeline is designed to solve text recognition tasks, extracting
 <tr>
 <td>PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>72.96</td>
-<td>5.87 / 5.87</td>
+<td>5.87 / 1.19</td>
 <td>9.07 / 4.28</td>
 <td>9.2 M</td>
 <td>PP-OCRv3’s lightweight recognition model is designed for high inference efficiency and can be deployed on a variety of hardware devices, including edge devices.</td>
@@ -201,15 +202,15 @@ The General OCR pipeline is designed to solve text recognition tasks, extracting
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Recognition Avg Accuracy(%)</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <tr>
 <td>ch_SVTRv2_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_SVTRv2_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_SVTRv2_rec_pretrained.pdparams">Training Model</a></td>
 <td>68.81</td>
-<td>8.08 / 8.08</td>
+<td>8.08 / 2.74</td>
 <td>50.17 / 42.50</td>
 <td>73.9 M</td>
 <td rowspan="1">
@@ -222,15 +223,15 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Recognition Avg Accuracy(%)</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <tr>
 <td>ch_RepSVTR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_RepSVTR_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_RepSVTR_rec_pretrained.pdparams">Training Model</a></td>
 <td>65.07</td>
-<td>5.93 / 5.93</td>
+<td>5.93 / 1.62</td>
 <td>20.73 / 7.32</td>
 <td>22.1 M</td>
 <td rowspan="1">    The RepSVTR text recognition model is a mobile text recognition model based on SVTRv2. It won the first prize in the PaddleOCR Algorithm Model Challenge - Task One: OCR End-to-End Recognition Task. The end-to-end recognition accuracy on the B list is 2.5% higher than that of PP-OCRv4, with the same inference speed.</td>
@@ -242,15 +243,15 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Recognition Avg Accuracy(%)</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <tr>
 <td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td> 70.39</td>
-<td>4.81 / 4.81</td>
+<td>4.81 / 0.75</td>
 <td>16.10 / 5.31</td>
 <td>6.8 M</td>
 <td>The ultra-lightweight English recognition model trained based on the PP-OCRv4 recognition model supports the recognition of English and numbers.</td>
@@ -258,7 +259,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>en_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>70.69</td>
-<td>5.44 / 5.44</td>
+<td>5.44 / 0.75</td>
 <td>8.65 / 5.57</td>
 <td>7.8 M </td>
 <td>The ultra-lightweight English recognition model trained based on the PP-OCRv3 recognition model supports the recognition of English and numbers.</td>
@@ -270,15 +271,15 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Recognition Avg Accuracy(%)</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <tr>
 <td>korean_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/korean_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>60.21</td>
-<td>5.40 / 5.40</td>
+<td>5.40 / 0.97</td>
 <td>9.11 / 4.05</td>
 <td>8.6 M</td>
 <td>The ultra-lightweight Korean recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Korean and numbers. </td>
@@ -286,7 +287,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>japan_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/japan_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>45.69</td>
-<td>5.70 / 5.70</td>
+<td>5.70 / 1.02</td>
 <td>8.48 / 4.07</td>
 <td>8.8 M </td>
 <td>The ultra-lightweight Japanese recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Japanese and numbers.</td>
@@ -294,7 +295,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>chinese_cht_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/chinese_cht_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>82.06</td>
-<td>5.90 / 5.90</td>
+<td>5.90 / 1.28</td>
 <td>9.28 / 4.34</td>
 <td>9.7 M </td>
 <td>The ultra-lightweight Traditional Chinese recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Traditional Chinese and numbers.</td>
@@ -302,7 +303,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>te_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/te_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>95.88</td>
-<td>5.42 / 5.42</td>
+<td>5.42 / 0.82</td>
 <td>8.10 / 6.91</td>
 <td>7.8 M </td>
 <td>The ultra-lightweight Telugu recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Telugu and numbers.</td>
@@ -310,7 +311,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>ka_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ka_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>96.96</td>
-<td>5.25 / 5.25</td>
+<td>5.25 / 0.79</td>
 <td>9.09 / 3.86</td>
 <td>8.0 M </td>
 <td>The ultra-lightweight Kannada recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Kannada and numbers.</td>
@@ -318,7 +319,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>ta_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ta_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>76.83</td>
-<td>5.23 / 5.23</td>
+<td>5.23 / 0.75</td>
 <td>10.13 / 4.30</td>
 <td>8.0 M </td>
 <td>The ultra-lightweight Tamil recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Tamil and numbers.</td>
@@ -326,7 +327,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>latin_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/latin_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>76.93</td>
-<td>5.20 / 5.20</td>
+<td>5.20 / 0.79</td>
 <td>8.83 / 7.15</td>
 <td>7.8 M</td>
 <td>The ultra-lightweight Latin recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Latin script and numbers.</td>
@@ -334,7 +335,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>arabic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/arabic_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>73.55</td>
-<td>5.35 / 5.35</td>
+<td>5.35 / 0.79</td>
 <td>8.80 / 4.56</td>
 <td>7.8 M</td>
 <td>The ultra-lightweight Arabic script recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Arabic script and numbers.</td>
@@ -342,7 +343,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <td>cyrillic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/cyrillic_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>94.28</td>
-<td>5.23 / 5.23</td>
+<td>5.23 / 0.76</td>
 <td>8.89 / 3.88</td>
 <td>7.9 M  </td>
 <td>
@@ -351,13 +352,12 @@ The ultra-lightweight cyrillic alphabet recognition model trained based on the P
 <tr>
 <td>devanagari_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/devanagari_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>96.44</td>
-<td>5.22 / 5.22</td>
+<td>5.22 / 0.79</td>
 <td>8.56 / 4.06</td>
 <td>7.9 M  </td>
 <td>The ultra-lightweight Devanagari script recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Devanagari script and numbers.</td>
 </tr>
 </table>
-
 </details>
 
 <p><b>Text Line Orientation Classification Module (Optional):</b></p>

docs/pipeline_usage/tutorials/ocr_pipelines/OCR.md

@@ -118,9 +118,9 @@ OCR（光学字符识别，Optical Character Recognition）是一种将图像中
 </tr>
 <tr>
 <td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_doc_pretrained.pdparams">训练模型</a></td>
 <td>81.53</td>
-<td>6.65 / 6.65</td>
+<td>6.65 / 2.38</td>
 <td>32.92 / 32.92</td>
 <td>74.7 M</td>
 <td>PP-OCRv4_server_rec_doc是在PP-OCRv4_server_rec的基础上，在更多中文文档数据和PP-OCR训练数据的混合数据训练而成，增加了部分繁体字、日文、特殊字符的识别能力，可支持识别的字符为1.5万+，除文档相关的文字识别能力提升外，也同时提升了通用文字的识别能力</td>
@@ -128,7 +128,7 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a><
 <tr>
 <td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>78.74</td>
-<td>4.82 / 4.82</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
 <td>PP-OCRv4的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
@@ -136,23 +136,23 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a><
 <tr>
 <td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">训练模型</a></td>
 <td>80.61 </td>
-<td>6.58 / 6.58</td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
 <td>PP-OCRv4的服务器端模型，推理精度高，可以部署在多种不同的服务器上</td>
 </tr>
 <tr>
 <td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>70.39</td>
-<td>4.81 / 4.81</td>
+<td>4.81 / 0.75</td>
 <td>16.10 / 5.31</td>
 <td>6.8 M</td>
 <td>基于PP-OCRv4识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
 </tr>
 </table>
 
-&gt;❗ 以上列出的是文本识别模块重点支持的<b>4个核心模型</b>，该模块总共支持<b>18个全量模型</b>，包含多个多语言文本识别模型，完整的模型列表如下：
+> ❗ 以上列出的是文本识别模块重点支持的<b>4个核心模型</b>，该模块总共支持<b>18个全量模型</b>，包含多个多语言文本识别模型，完整的模型列表如下：
 
 <details><summary> 👉模型列表详情</summary>
 
@@ -168,9 +168,9 @@ en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></
 </tr>
 <tr>
 <td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_doc_pretrained.pdparams">训练模型</a></td>
 <td>81.53</td>
-<td>6.65 / 6.65</td>
+<td>6.65 / 2.38</td>
 <td>32.92 / 32.92</td>
 <td>74.7 M</td>
 <td>PP-OCRv4_server_rec_doc是在PP-OCRv4_server_rec的基础上，在更多中文文档数据和PP-OCR训练数据的混合数据训练而成，增加了部分繁体字、日文、特殊字符的识别能力，可支持识别的字符为1.5万+，除文档相关的文字识别能力提升外，也同时提升了通用文字的识别能力</td>
@@ -178,7 +178,7 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a><
 <tr>
 <td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>78.74</td>
-<td>4.82 / 4.82</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
 <td>PP-OCRv4的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
@@ -186,16 +186,16 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a><
 <tr>
 <td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">训练模型</a></td>
 <td>80.61 </td>
-<td>6.58 / 6.58</td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
 <td>PP-OCRv4的服务器端模型，推理精度高，可以部署在多种不同的服务器上</td>
 </tr>
 <tr>
 <td>PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>72.96</td>
-<td>5.87 / 5.87</td>
+<td>5.87 / 1.19</td>
 <td>9.07 / 4.28</td>
 <td>9.2 M</td>
 <td>PP-OCRv3的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
@@ -214,7 +214,7 @@ PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
 <tr>
 <td>ch_SVTRv2_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_SVTRv2_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_SVTRv2_rec_pretrained.pdparams">训练模型</a></td>
 <td>68.81</td>
-<td>8.08 / 8.08</td>
+<td>8.08 / 2.74</td>
 <td>50.17 / 42.50</td>
 <td>73.9 M</td>
 <td rowspan="1">
@@ -235,7 +235,7 @@ SVTRv2 是一种由复旦大学视觉与学习实验室（FVL）的OpenOCR团队
 <tr>
 <td>ch_RepSVTR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_RepSVTR_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_RepSVTR_rec_pretrained.pdparams">训练模型</a></td>
 <td>65.07</td>
-<td>5.93 / 5.93</td>
+<td>5.93 / 1.62</td>
 <td>20.73 / 7.32</td>
 <td>22.1 M</td>
 <td rowspan="1">    RepSVTR 文本识别模型是一种基于SVTRv2 的移动端文本识别模型，其在PaddleOCR算法模型挑战赛 - 赛题一：OCR端到端识别任务中荣获一等奖，B榜端到端识别精度相比PP-OCRv4提升2.5%，推理速度持平。</td>
@@ -254,18 +254,18 @@ SVTRv2 是一种由复旦大学视觉与学习实验室（FVL）的OpenOCR团队
 </tr>
 <tr>
 <td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td> 70.39</td>
-<td>4.81 / 4.81</td>
+<td>4.81 / 0.75</td>
 <td>16.10 / 5.31</td>
 <td>6.8 M</td>
 <td>基于PP-OCRv4识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
 </tr>
 <tr>
 <td>en_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-en_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+en_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>70.69</td>
-<td>5.44 / 5.44</td>
+<td>5.44 / 0.75</td>
 <td>8.65 / 5.57</td>
 <td>7.8 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
@@ -284,90 +284,90 @@ en_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></
 </tr>
 <tr>
 <td>korean_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-korean_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+korean_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/korean_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>60.21</td>
-<td>5.40 / 5.40</td>
+<td>5.40 / 0.97</td>
 <td>9.11 / 4.05</td>
 <td>8.6 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量韩文识别模型，支持韩文、数字识别</td>
 </tr>
 <tr>
 <td>japan_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-japan_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+japan_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/japan_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>45.69</td>
-<td>5.70 / 5.70</td>
+<td>5.70 / 1.02</td>
 <td>8.48 / 4.07</td>
 <td>8.8 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量日文识别模型，支持日文、数字识别</td>
 </tr>
 <tr>
 <td>chinese_cht_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-chinese_cht_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+chinese_cht_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/chinese_cht_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>82.06</td>
-<td>5.90 / 5.90</td>
+<td>5.90 / 1.28</td>
 <td>9.28 / 4.34</td>
 <td>9.7 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量繁体中文识别模型，支持繁体中文、数字识别</td>
 </tr>
 <tr>
 <td>te_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-te_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+te_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/te_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>95.88</td>
-<td>5.42 / 5.42</td>
+<td>5.42 / 0.82</td>
 <td>8.10 / 6.91</td>
 <td>7.8 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量泰卢固文识别模型，支持泰卢固文、数字识别</td>
 </tr>
 <tr>
 <td>ka_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-ka_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+ka_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ka_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>96.96</td>
-<td>5.25 / 5.25</td>
+<td>5.25 / 0.79</td>
 <td>9.09 / 3.86</td>
 <td>8.0 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量卡纳达文识别模型，支持卡纳达文、数字识别</td>
 </tr>
 <tr>
 <td>ta_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-ta_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+ta_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ta_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>76.83</td>
-<td>5.23 / 5.23</td>
+<td>5.23 / 0.75</td>
 <td>10.13 / 4.30</td>
 <td>8.0 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量泰米尔文识别模型，支持泰米尔文、数字识别</td>
 </tr>
 <tr>
 <td>latin_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-latin_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+latin_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/latin_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>76.93</td>
-<td>5.20 / 5.20</td>
+<td>5.20 / 0.79</td>
 <td>8.83 / 7.15</td>
 <td>7.8 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量拉丁文识别模型，支持拉丁文、数字识别</td>
 </tr>
 <tr>
 <td>arabic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-arabic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+arabic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/arabic_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>73.55</td>
-<td>5.35 / 5.35</td>
+<td>5.35 / 0.79</td>
 <td>8.80 / 4.56</td>
 <td>7.8 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量阿拉伯字母识别模型，支持阿拉伯字母、数字识别</td>
 </tr>
 <tr>
 <td>cyrillic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-cyrillic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+cyrillic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/cyrillic_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>94.28</td>
-<td>5.23 / 5.23</td>
+<td>5.23 / 0.76</td>
 <td>8.89 / 3.88</td>
 <td>7.9 M  </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量斯拉夫字母识别模型，支持斯拉夫字母、数字识别</td>
 </tr>
 <tr>
 <td>devanagari_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-devanagari_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+devanagari_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/devanagari_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>96.44</td>
-<td>5.22 / 5.22</td>
+<td>5.22 / 0.79</td>
 <td>8.56 / 4.06</td>
 <td>7.9 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量梵文字母识别模型，支持梵文字母、数字识别</td>

docs/pipeline_usage/tutorials/ocr_pipelines/doc_preprocessor.en.md

@@ -19,7 +19,7 @@ The document image preprocessing pipeline integrates two major functions: docume
 <tr>
 <th>Model</th><th>Model download link</th>
 <th>Top-1 Acc（%）</th>
-<th>GPU inference time (ms)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU inference time (ms)</th>
 <th>Model storage size（M)</th>
 <th>Introduction</th>

docs/pipeline_usage/tutorials/ocr_pipelines/layout_parsing.en.md

@@ -243,49 +243,67 @@ The <b>General Layout Parsing Pipeline</b> includes modules for table structure
 </table>
 
 <p><b>Text Recognition Module Models</b>:</p>
+* <b>Chinese Recognition Model</b>
 <table>
 <tr>
 <th>Model</th><th>Model Download Link</th>
-<th>Recognition Avg Accuracy (%)</th>
+<th>Recognition Avg Accuracy(%)</th>
 <th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
-<th>Model Size (M)</th>
-<th>Description</th>
+<th>Model Storage Size (M)</th>
+<th>Introduction</th>
 </tr>
 <tr>
-<td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Trained Model</a></td>
-<td>78.20</td>
-<td>4.82 / 4.82</td>
+<td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_doc_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>81.53</td>
+<td>6.65 / 2.38</td>
+<td>32.92 / 32.92</td>
+<td>74.7 M</td>
+<td>PP-OCRv4_server_rec_doc is trained on a mixed dataset of more Chinese document data and PP-OCR training data based on PP-OCRv4_server_rec. It has added the recognition capabilities for some traditional Chinese characters, Japanese, and special characters. The number of recognizable characters is over 15,000. In addition to the improvement in document-related text recognition, it also enhances the general text recognition capability.</td>
+</tr>
+<tr>
+<td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Training Model</a></td>
+<td>78.74</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
-<td rowspan="2">PP-OCRv4 is the next version of Baidu PaddlePaddle's self-developed text recognition model PP-OCRv3. By introducing data augmentation schemes and GTC-NRTR guidance branches, it further improves text recognition accuracy without compromising inference speed. The model offers both server (server) and mobile (mobile) versions to meet industrial needs in different scenarios.</td>
+<td>The lightweight recognition model of PP-OCRv4 has high inference efficiency and can be deployed on various hardware devices, including edge devices.</td>
 </tr>
 <tr>
-<td>PP-OCRv4_server_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Trained Model</a></td>
-<td>79.20</td>
-<td>6.58 / 6.58</td>
+<td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Trained Model</a></td>
+<td>80.61 </td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
+<td>The server-side model of PP-OCRv4 offers high inference accuracy and can be deployed on various types of servers.</td>
+</tr>
+<tr>
+<td>PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>72.96</td>
+<td>5.87 / 1.19</td>
+<td>9.07 / 4.28</td>
+<td>9.2 M</td>
+<td>PP-OCRv3’s lightweight recognition model is designed for high inference efficiency and can be deployed on a variety of hardware devices, including edge devices.</td>
 </tr>
 </table>
 
 <table>
 <tr>
 <th>Model</th><th>Model Download Link</th>
-<th>Recognition Avg Accuracy (%)</th>
+<th>Recognition Avg Accuracy(%)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
-<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
-<th>Model Size (M)</th>
-<th>Description</th>
+<th>Model Storage Size (M)</th>
+<th>Introduction</th>
 </tr>
 <tr>
-<td>ch_SVTRv2_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_SVTRv2_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_SVTRv2_rec_pretrained.pdparams">Trained Model</a></td>
+<td>ch_SVTRv2_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_SVTRv2_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_SVTRv2_rec_pretrained.pdparams">Training Model</a></td>
 <td>68.81</td>
-<td>8.08 / 8.08</td>
+<td>8.08 / 2.74</td>
 <td>50.17 / 42.50</td>
 <td>73.9 M</td>
 <td rowspan="1">
-SVTRv2 is a server-side text recognition model developed by the OpenOCR team at the Vision and Learning Lab (FVL) of Fudan University. It won the first prize in the OCR End-to-End Recognition Task of the PaddleOCR Algorithm Model Challenge, with a 6% improvement in end-to-end recognition accuracy compared to PP-OCRv4 on the A-list.
+SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan University's Visual and Learning Laboratory (FVL). It won the first prize in the PaddleOCR Algorithm Model Challenge - Task One: OCR End-to-End Recognition Task. The end-to-end recognition accuracy on the A list is 6% higher than that of PP-OCRv4.
 </td>
 </tr>
 </table>
@@ -293,21 +311,140 @@ SVTRv2 is a server-side text recognition model developed by the OpenOCR team at
 <table>
 <tr>
 <th>Model</th><th>Model Download Link</th>
-<th>Recognition Avg Accuracy (%)</th>
-<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>Recognition Avg Accuracy(%)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
-<th>Model Size (M)</th>
-<th>Description</th>
+<th>Model Storage Size (M)</th>
+<th>Introduction</th>
 </tr>
 <tr>
-<td>ch_RepSVTR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_RepSVTR_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_RepSVTR_rec_pretrained.pdparams">Trained Model</a></td>
+<td>ch_RepSVTR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_RepSVTR_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_RepSVTR_rec_pretrained.pdparams">Training Model</a></td>
 <td>65.07</td>
-<td>5.93 / 5.93</td>
+<td>5.93 / 1.62</td>
 <td>20.73 / 7.32</td>
 <td>22.1 M</td>
-<td rowspan="1">
-The RepSVTR text recognition model is a mobile-oriented text recognition model based on SVTRv2. It won the first prize in the OCR End-to-End Recognition Task of the PaddleOCR Algorithm Model Challenge, with a 2.5% improvement in end-to-end recognition accuracy compared to PP-OCRv4 on the B-list, while maintaining similar inference speed.
-</td>
+<td rowspan="1">    The RepSVTR text recognition model is a mobile text recognition model based on SVTRv2. It won the first prize in the PaddleOCR Algorithm Model Challenge - Task One: OCR End-to-End Recognition Task. The end-to-end recognition accuracy on the B list is 2.5% higher than that of PP-OCRv4, with the same inference speed.</td>
+</tr>
+</table>
+
+* <b>English Recognition Model</b>
+<table>
+<tr>
+<th>Model</th><th>Model Download Link</th>
+<th>Recognition Avg Accuracy(%)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>Model Storage Size (M)</th>
+<th>Introduction</th>
+</tr>
+<tr>
+<td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td> 70.39</td>
+<td>4.81 / 0.75</td>
+<td>16.10 / 5.31</td>
+<td>6.8 M</td>
+<td>The ultra-lightweight English recognition model trained based on the PP-OCRv4 recognition model supports the recognition of English and numbers.</td>
+</tr>
+<tr>
+<td>en_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>70.69</td>
+<td>5.44 / 0.75</td>
+<td>8.65 / 5.57</td>
+<td>7.8 M </td>
+<td>The ultra-lightweight English recognition model trained based on the PP-OCRv3 recognition model supports the recognition of English and numbers.</td>
+</tr>
+</table>
+
+* <b>Multilingual Recognition Model</b>
+<table>
+<tr>
+<th>Model</th><th>Model Download Link</th>
+<th>Recognition Avg Accuracy(%)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>Model Storage Size (M)</th>
+<th>Introduction</th>
+</tr>
+<tr>
+<td>korean_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/korean_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>60.21</td>
+<td>5.40 / 0.97</td>
+<td>9.11 / 4.05</td>
+<td>8.6 M</td>
+<td>The ultra-lightweight Korean recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Korean and numbers. </td>
+</tr>
+<tr>
+<td>japan_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/japan_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>45.69</td>
+<td>5.70 / 1.02</td>
+<td>8.48 / 4.07</td>
+<td>8.8 M </td>
+<td>The ultra-lightweight Japanese recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Japanese and numbers.</td>
+</tr>
+<tr>
+<td>chinese_cht_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/chinese_cht_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>82.06</td>
+<td>5.90 / 1.28</td>
+<td>9.28 / 4.34</td>
+<td>9.7 M </td>
+<td>The ultra-lightweight Traditional Chinese recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Traditional Chinese and numbers.</td>
+</tr>
+<tr>
+<td>te_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/te_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>95.88</td>
+<td>5.42 / 0.82</td>
+<td>8.10 / 6.91</td>
+<td>7.8 M </td>
+<td>The ultra-lightweight Telugu recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Telugu and numbers.</td>
+</tr>
+<tr>
+<td>ka_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ka_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>96.96</td>
+<td>5.25 / 0.79</td>
+<td>9.09 / 3.86</td>
+<td>8.0 M </td>
+<td>The ultra-lightweight Kannada recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Kannada and numbers.</td>
+</tr>
+<tr>
+<td>ta_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ta_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>76.83</td>
+<td>5.23 / 0.75</td>
+<td>10.13 / 4.30</td>
+<td>8.0 M </td>
+<td>The ultra-lightweight Tamil recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Tamil and numbers.</td>
+</tr>
+<tr>
+<td>latin_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/latin_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>76.93</td>
+<td>5.20 / 0.79</td>
+<td>8.83 / 7.15</td>
+<td>7.8 M</td>
+<td>The ultra-lightweight Latin recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Latin script and numbers.</td>
+</tr>
+<tr>
+<td>arabic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/arabic_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>73.55</td>
+<td>5.35 / 0.79</td>
+<td>8.80 / 4.56</td>
+<td>7.8 M</td>
+<td>The ultra-lightweight Arabic script recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Arabic script and numbers.</td>
+</tr>
+<tr>
+<td>cyrillic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/cyrillic_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>94.28</td>
+<td>5.23 / 0.76</td>
+<td>8.89 / 3.88</td>
+<td>7.9 M  </td>
+<td>
+The ultra-lightweight cyrillic alphabet recognition model trained based on the PP-OCRv3 recognition model supports the recognition of cyrillic letters and numbers.</td>
+</tr>
+<tr>
+<td>devanagari_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/devanagari_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>96.44</td>
+<td>5.22 / 0.79</td>
+<td>8.56 / 4.06</td>
+<td>7.9 M  </td>
+<td>The ultra-lightweight Devanagari script recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Devanagari script and numbers.</td>
 </tr>
 </table>
 

docs/pipeline_usage/tutorials/ocr_pipelines/layout_parsing.md

@@ -248,9 +248,9 @@ comments: true
 </tr>
 <tr>
 <td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_doc_pretrained.pdparams">训练模型</a></td>
 <td>81.53</td>
-<td>6.65 / 6.65</td>
+<td>6.65 / 2.38</td>
 <td>32.92 / 32.92</td>
 <td>74.7 M</td>
 <td>PP-OCRv4_server_rec_doc是在PP-OCRv4_server_rec的基础上，在更多中文文档数据和PP-OCR训练数据的混合数据训练而成，增加了部分繁体字、日文、特殊字符的识别能力，可支持识别的字符为1.5万+，除文档相关的文字识别能力提升外，也同时提升了通用文字的识别能力</td>
@@ -258,7 +258,7 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a><
 <tr>
 <td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>78.74</td>
-<td>4.82 / 4.82</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
 <td>PP-OCRv4的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
@@ -266,16 +266,16 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a><
 <tr>
 <td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">训练模型</a></td>
 <td>80.61 </td>
-<td>6.58 / 6.58</td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
 <td>PP-OCRv4的服务器端模型，推理精度高，可以部署在多种不同的服务器上</td>
 </tr>
 <tr>
 <td>PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>72.96</td>
-<td>5.87 / 5.87</td>
+<td>5.87 / 1.19</td>
 <td>9.07 / 4.28</td>
 <td>9.2 M</td>
 <td>PP-OCRv3的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
@@ -294,7 +294,7 @@ PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
 <tr>
 <td>ch_SVTRv2_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_SVTRv2_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_SVTRv2_rec_pretrained.pdparams">训练模型</a></td>
 <td>68.81</td>
-<td>8.08 / 8.08</td>
+<td>8.08 / 2.74</td>
 <td>50.17 / 42.50</td>
 <td>73.9 M</td>
 <td rowspan="1">
@@ -315,7 +315,7 @@ SVTRv2 是一种由复旦大学视觉与学习实验室（FVL）的OpenOCR团队
 <tr>
 <td>ch_RepSVTR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_RepSVTR_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_RepSVTR_rec_pretrained.pdparams">训练模型</a></td>
 <td>65.07</td>
-<td>5.93 / 5.93</td>
+<td>5.93 / 1.62</td>
 <td>20.73 / 7.32</td>
 <td>22.1 M</td>
 <td rowspan="1">    RepSVTR 文本识别模型是一种基于SVTRv2 的移动端文本识别模型，其在PaddleOCR算法模型挑战赛 - 赛题一：OCR端到端识别任务中荣获一等奖，B榜端到端识别精度相比PP-OCRv4提升2.5%，推理速度持平。</td>
@@ -335,18 +335,18 @@ SVTRv2 是一种由复旦大学视觉与学习实验室（FVL）的OpenOCR团队
 </tr>
 <tr>
 <td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td> 70.39</td>
-<td>4.81 / 4.81</td>
+<td>4.81 / 0.75</td>
 <td>16.10 / 5.31</td>
 <td>6.8 M</td>
 <td>基于PP-OCRv4识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
 </tr>
 <tr>
 <td>en_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-en_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+en_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>70.69</td>
-<td>5.44 / 5.44</td>
+<td>5.44 / 0.75</td>
 <td>8.65 / 5.57</td>
 <td>7.8 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
@@ -365,90 +365,90 @@ en_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></
 </tr>
 <tr>
 <td>korean_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-korean_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+korean_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/korean_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>60.21</td>
-<td>5.40 / 5.40</td>
+<td>5.40 / 0.97</td>
 <td>9.11 / 4.05</td>
 <td>8.6 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量韩文识别模型，支持韩文、数字识别</td>
 </tr>
 <tr>
 <td>japan_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-japan_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+japan_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/japan_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>45.69</td>
-<td>5.70 / 5.70</td>
+<td>5.70 / 1.02</td>
 <td>8.48 / 4.07</td>
 <td>8.8 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量日文识别模型，支持日文、数字识别</td>
 </tr>
 <tr>
 <td>chinese_cht_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-chinese_cht_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+chinese_cht_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/chinese_cht_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>82.06</td>
-<td>5.90 / 5.90</td>
+<td>5.90 / 1.28</td>
 <td>9.28 / 4.34</td>
 <td>9.7 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量繁体中文识别模型，支持繁体中文、数字识别</td>
 </tr>
 <tr>
 <td>te_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-te_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+te_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/te_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>95.88</td>
-<td>5.42 / 5.42</td>
+<td>5.42 / 0.82</td>
 <td>8.10 / 6.91</td>
 <td>7.8 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量泰卢固文识别模型，支持泰卢固文、数字识别</td>
 </tr>
 <tr>
 <td>ka_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-ka_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+ka_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ka_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>96.96</td>
-<td>5.25 / 5.25</td>
+<td>5.25 / 0.79</td>
 <td>9.09 / 3.86</td>
 <td>8.0 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量卡纳达文识别模型，支持卡纳达文、数字识别</td>
 </tr>
 <tr>
 <td>ta_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-ta_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+ta_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ta_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>76.83</td>
-<td>5.23 / 5.23</td>
+<td>5.23 / 0.75</td>
 <td>10.13 / 4.30</td>
 <td>8.0 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量泰米尔文识别模型，支持泰米尔文、数字识别</td>
 </tr>
 <tr>
 <td>latin_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-latin_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+latin_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/latin_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>76.93</td>
-<td>5.20 / 5.20</td>
+<td>5.20 / 0.79</td>
 <td>8.83 / 7.15</td>
 <td>7.8 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量拉丁文识别模型，支持拉丁文、数字识别</td>
 </tr>
 <tr>
 <td>arabic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-arabic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+arabic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/arabic_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>73.55</td>
-<td>5.35 / 5.35</td>
+<td>5.35 / 0.79</td>
 <td>8.80 / 4.56</td>
 <td>7.8 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量阿拉伯字母识别模型，支持阿拉伯字母、数字识别</td>
 </tr>
 <tr>
 <td>cyrillic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-cyrillic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+cyrillic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/cyrillic_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>94.28</td>
-<td>5.23 / 5.23</td>
+<td>5.23 / 0.76</td>
 <td>8.89 / 3.88</td>
 <td>7.9 M  </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量斯拉夫字母识别模型，支持斯拉夫字母、数字识别</td>
 </tr>
 <tr>
 <td>devanagari_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-devanagari_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+devanagari_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/devanagari_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>96.44</td>
-<td>5.22 / 5.22</td>
+<td>5.22 / 0.79</td>
 <td>8.56 / 4.06</td>
 <td>7.9 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量梵文字母识别模型，支持梵文字母、数字识别</td>

docs/pipeline_usage/tutorials/ocr_pipelines/layout_parsing_v2.en.md

@@ -237,7 +237,7 @@ Layout analysis is a technology that extracts structured information from docume
 
 <p><b>Text Recognition Module Model (Required):</b></p>
 
-* <b>Chinese Text Recognition Model</b>
+* <b>Chinese Recognition Model</b>
 <table>
 <tr>
 <th>Model</th><th>Model Download Link</th>
@@ -250,34 +250,34 @@ Layout analysis is a technology that extracts structured information from docume
 <tr>
 <td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_doc_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>81.53</td>
-<td>6.65 / 6.65</td>
+<td>6.65 / 2.38</td>
 <td>32.92 / 32.92</td>
 <td>74.7 M</td>
-<td>PP-OCRv4_server_rec_doc is trained on a mixed dataset of more Chinese document data and PP-OCR training data, based on PP-OCRv4_server_rec. It enhances the recognition capability of traditional Chinese characters, Japanese characters, and special characters, supporting over 15,000 characters. In addition to improved document-related text recognition, it also enhances general text recognition capabilities.</td>
+<td>PP-OCRv4_server_rec_doc is trained on a mixed dataset of more Chinese document data and PP-OCR training data based on PP-OCRv4_server_rec. It has added the recognition capabilities for some traditional Chinese characters, Japanese, and special characters. The number of recognizable characters is over 15,000. In addition to the improvement in document-related text recognition, it also enhances the general text recognition capability.</td>
 </tr>
 <tr>
 <td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Training Model</a></td>
 <td>78.74</td>
-<td>4.82 / 4.82</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
-<td>The lightweight recognition model of PP-OCRv4, with high inference efficiency, can be deployed on various hardware devices, including edge devices.</td>
+<td>The lightweight recognition model of PP-OCRv4 has high inference efficiency and can be deployed on various hardware devices, including edge devices.</td>
 </tr>
 <tr>
-<td>PP-OCRv4_server_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Training Model</a></td>
-<td>80.61</td>
-<td>6.58 / 6.58</td>
+<td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Trained Model</a></td>
+<td>80.61 </td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
-<td>The server-side model of PP-OCRv4, with high inference accuracy, can be deployed on various servers.</td>
+<td>The server-side model of PP-OCRv4 offers high inference accuracy and can be deployed on various types of servers.</td>
 </tr>
 <tr>
 <td>PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>72.96</td>
-<td>5.87 / 5.87</td>
+<td>5.87 / 1.19</td>
 <td>9.07 / 4.28</td>
 <td>9.2 M</td>
-<td>The lightweight recognition model of PP-OCRv3, with high inference efficiency, can be deployed on various hardware devices, including edge devices.</td>
+<td>PP-OCRv3’s lightweight recognition model is designed for high inference efficiency and can be deployed on a variety of hardware devices, including edge devices.</td>
 </tr>
 </table>
 
@@ -285,19 +285,19 @@ Layout analysis is a technology that extracts structured information from docume
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Recognition Avg Accuracy(%)</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <tr>
 <td>ch_SVTRv2_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_SVTRv2_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_SVTRv2_rec_pretrained.pdparams">Training Model</a></td>
 <td>68.81</td>
-<td>8.36801</td>
-<td>165.706</td>
+<td>8.08 / 2.74</td>
+<td>50.17 / 42.50</td>
 <td>73.9 M</td>
 <td rowspan="1">
-SVTRv2 is a server-side text recognition model developed by the OpenOCR team from the Visual and Learning Laboratory (FVL) at Fudan University. It won the first prize in the PaddleOCR Algorithm Model Challenge - Task 1: OCR End-to-End Recognition Task, with a 6% improvement in end-to-end recognition accuracy compared to PP-OCRv4.
+SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan University's Visual and Learning Laboratory (FVL). It won the first prize in the PaddleOCR Algorithm Model Challenge - Task One: OCR End-to-End Recognition Task. The end-to-end recognition accuracy on the A list is 6% higher than that of PP-OCRv4.
 </td>
 </tr>
 </table>
@@ -306,18 +306,18 @@ SVTRv2 is a server-side text recognition model developed by the OpenOCR team fro
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Recognition Avg Accuracy(%)</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <tr>
 <td>ch_RepSVTR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_RepSVTR_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_RepSVTR_rec_pretrained.pdparams">Training Model</a></td>
 <td>65.07</td>
-<td>10.5047</td>
-<td>51.5647</td>
+<td>5.93 / 1.62</td>
+<td>20.73 / 7.32</td>
 <td>22.1 M</td>
-<td rowspan="1">RepSVTR is a mobile text recognition model based on SVTRv2. It won the first prize in the PaddleOCR Algorithm Model Challenge - Task 1: OCR End-to-End Recognition Task, with a 2.5% improvement in end-to-end recognition accuracy compared to PP-OCRv4 and comparable inference speed.</td>
+<td rowspan="1">    The RepSVTR text recognition model is a mobile text recognition model based on SVTRv2. It won the first prize in the PaddleOCR Algorithm Model Challenge - Task One: OCR End-to-End Recognition Task. The end-to-end recognition accuracy on the B list is 2.5% higher than that of PP-OCRv4, with the same inference speed.</td>
 </tr>
 </table>
 
@@ -326,117 +326,119 @@ SVTRv2 is a server-side text recognition model developed by the OpenOCR team fro
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Recognition Avg Accuracy(%)</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <tr>
 <td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td> 70.39</td>
-<td></td>
-<td></td>
+<td>4.81 / 0.75</td>
+<td>16.10 / 5.31</td>
 <td>6.8 M</td>
-<td>An ultra-lightweight English recognition model trained based on the PP-OCRv4 recognition model, supporting English and number recognition.</td>
+<td>The ultra-lightweight English recognition model trained based on the PP-OCRv4 recognition model supports the recognition of English and numbers.</td>
 </tr>
 <tr>
 <td>en_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>70.69</td>
-<td></td>
-<td></td>
+<td>5.44 / 0.75</td>
+<td>8.65 / 5.57</td>
 <td>7.8 M </td>
-<td>An ultra-lightweight English recognition model trained based on the PP-OCRv3 recognition model, supporting English and number recognition.</td>
+<td>The ultra-lightweight English recognition model trained based on the PP-OCRv3 recognition model supports the recognition of English and numbers.</td>
 </tr>
 </table>
-<b>Multilingual Recognition Models</b>
+
+* <b>Multilingual Recognition Model</b>
 <table>
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Recognition Avg Accuracy(%)</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <tr>
 <td>korean_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/korean_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>60.21</td>
-<td></td>
-<td></td>
+<td>5.40 / 0.97</td>
+<td>9.11 / 4.05</td>
 <td>8.6 M</td>
-<td>An ultra-lightweight Korean recognition model trained based on the PP-OCRv3 recognition model, supporting recognition of Korean and numbers.</td>
+<td>The ultra-lightweight Korean recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Korean and numbers. </td>
 </tr>
 <tr>
 <td>japan_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/japan_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>45.69</td>
-<td></td>
-<td></td>
-<td>8.8 M</td>
-<td>An ultra-lightweight Japanese recognition model trained based on the PP-OCRv3 recognition model, supporting recognition of Japanese and numbers.</td>
+<td>5.70 / 1.02</td>
+<td>8.48 / 4.07</td>
+<td>8.8 M </td>
+<td>The ultra-lightweight Japanese recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Japanese and numbers.</td>
 </tr>
 <tr>
 <td>chinese_cht_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/chinese_cht_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>82.06</td>
-<td></td>
-<td></td>
-<td>9.7 M</td>
-<td>An ultra-lightweight Traditional Chinese recognition model trained based on the PP-OCRv3 recognition model, supporting recognition of Traditional Chinese and numbers.</td>
+<td>5.90 / 1.28</td>
+<td>9.28 / 4.34</td>
+<td>9.7 M </td>
+<td>The ultra-lightweight Traditional Chinese recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Traditional Chinese and numbers.</td>
 </tr>
 <tr>
 <td>te_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/te_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>95.88</td>
-<td></td>
-<td></td>
-<td>7.8 M</td>
-<td>An ultra-lightweight Telugu recognition model trained based on the PP-OCRv3 recognition model, supporting recognition of Telugu and numbers.</td>
+<td>5.42 / 0.82</td>
+<td>8.10 / 6.91</td>
+<td>7.8 M </td>
+<td>The ultra-lightweight Telugu recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Telugu and numbers.</td>
 </tr>
 <tr>
 <td>ka_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ka_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>96.96</td>
-<td></td>
-<td></td>
-<td>8.0 M</td>
-<td>An ultra-lightweight Kannada recognition model trained based on the PP-OCRv3 recognition model, supporting recognition of Kannada and numbers.</td>
+<td>5.25 / 0.79</td>
+<td>9.09 / 3.86</td>
+<td>8.0 M </td>
+<td>The ultra-lightweight Kannada recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Kannada and numbers.</td>
 </tr>
 <tr>
 <td>ta_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ta_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>76.83</td>
-<td></td>
-<td></td>
-<td>8.0 M</td>
-<td>An ultra-lightweight Tamil recognition model trained based on the PP-OCRv3 recognition model, supporting recognition of Tamil and numbers.</td>
+<td>5.23 / 0.75</td>
+<td>10.13 / 4.30</td>
+<td>8.0 M </td>
+<td>The ultra-lightweight Tamil recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Tamil and numbers.</td>
 </tr>
 <tr>
 <td>latin_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/latin_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>76.93</td>
-<td></td>
-<td></td>
+<td>5.20 / 0.79</td>
+<td>8.83 / 7.15</td>
 <td>7.8 M</td>
-<td>An ultra-lightweight Latin recognition model trained based on the PP-OCRv3 recognition model, supporting recognition of Latin and numbers.</td>
+<td>The ultra-lightweight Latin recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Latin script and numbers.</td>
 </tr>
 <tr>
 <td>arabic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/arabic_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>73.55</td>
-<td></td>
-<td></td>
+<td>5.35 / 0.79</td>
+<td>8.80 / 4.56</td>
 <td>7.8 M</td>
-<td>An ultra-lightweight Arabic letter recognition model trained based on the PP-OCRv3 recognition model, supporting recognition of Arabic letters and numbers.</td>
+<td>The ultra-lightweight Arabic script recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Arabic script and numbers.</td>
 </tr>
 <tr>
 <td>cyrillic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/cyrillic_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>94.28</td>
-<td></td>
-<td></td>
-<td>7.9 M</td>
-<td>An ultra-lightweight Cyrillic letter recognition model trained based on the PP-OCRv3 recognition model, supporting recognition of Cyrillic letters and numbers.</td>
+<td>5.23 / 0.76</td>
+<td>8.89 / 3.88</td>
+<td>7.9 M  </td>
+<td>
+The ultra-lightweight cyrillic alphabet recognition model trained based on the PP-OCRv3 recognition model supports the recognition of cyrillic letters and numbers.</td>
 </tr>
 <tr>
 <td>devanagari_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/devanagari_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
 <td>96.44</td>
-<td></td>
-<td></td>
-<td>7.9 M</td>
-<td>An ultra-lightweight Devanagari letter recognition model trained based on the PP-OCRv3 recognition model, supporting recognition of Devanagari letters and numbers.</td>
+<td>5.22 / 0.79</td>
+<td>8.56 / 4.06</td>
+<td>7.9 M  </td>
+<td>The ultra-lightweight Devanagari script recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Devanagari script and numbers.</td>
 </tr>
 </table>
 

docs/pipeline_usage/tutorials/ocr_pipelines/layout_parsing_v2.md

@@ -183,9 +183,9 @@ comments: true
 </tr>
 <tr>
 <td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_doc_pretrained.pdparams">训练模型</a></td>
 <td>81.53</td>
-<td>6.65 / 6.65</td>
+<td>6.65 / 2.38</td>
 <td>32.92 / 32.92</td>
 <td>74.7 M</td>
 <td>PP-OCRv4_server_rec_doc是在PP-OCRv4_server_rec的基础上，在更多中文文档数据和PP-OCR训练数据的混合数据训练而成，增加了部分繁体字、日文、特殊字符的识别能力，可支持识别的字符为1.5万+，除文档相关的文字识别能力提升外，也同时提升了通用文字的识别能力</td>
@@ -193,7 +193,7 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a><
 <tr>
 <td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>78.74</td>
-<td>4.82 / 4.82</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
 <td>PP-OCRv4的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
@@ -201,16 +201,16 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a><
 <tr>
 <td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">训练模型</a></td>
 <td>80.61 </td>
-<td>6.58 / 6.58</td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
 <td>PP-OCRv4的服务器端模型，推理精度高，可以部署在多种不同的服务器上</td>
 </tr>
 <tr>
 <td>PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>72.96</td>
-<td>5.87 / 5.87</td>
+<td>5.87 / 1.19</td>
 <td>9.07 / 4.28</td>
 <td>9.2 M</td>
 <td>PP-OCRv3的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
@@ -229,7 +229,7 @@ PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
 <tr>
 <td>ch_SVTRv2_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_SVTRv2_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_SVTRv2_rec_pretrained.pdparams">训练模型</a></td>
 <td>68.81</td>
-<td>8.08 / 8.08</td>
+<td>8.08 / 2.74</td>
 <td>50.17 / 42.50</td>
 <td>73.9 M</td>
 <td rowspan="1">
@@ -250,7 +250,7 @@ SVTRv2 是一种由复旦大学视觉与学习实验室（FVL）的OpenOCR团队
 <tr>
 <td>ch_RepSVTR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_RepSVTR_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_RepSVTR_rec_pretrained.pdparams">训练模型</a></td>
 <td>65.07</td>
-<td>5.93 / 5.93</td>
+<td>5.93 / 1.62</td>
 <td>20.73 / 7.32</td>
 <td>22.1 M</td>
 <td rowspan="1">    RepSVTR 文本识别模型是一种基于SVTRv2 的移动端文本识别模型，其在PaddleOCR算法模型挑战赛 - 赛题一：OCR端到端识别任务中荣获一等奖，B榜端到端识别精度相比PP-OCRv4提升2.5%，推理速度持平。</td>
@@ -269,18 +269,18 @@ SVTRv2 是一种由复旦大学视觉与学习实验室（FVL）的OpenOCR团队
 </tr>
 <tr>
 <td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td> 70.39</td>
-<td>4.81 / 4.81</td>
+<td>4.81 / 0.75</td>
 <td>16.10 / 5.31</td>
 <td>6.8 M</td>
 <td>基于PP-OCRv4识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
 </tr>
 <tr>
 <td>en_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-en_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+en_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>70.69</td>
-<td>5.44 / 5.44</td>
+<td>5.44 / 0.75</td>
 <td>8.65 / 5.57</td>
 <td>7.8 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
@@ -299,90 +299,90 @@ en_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></
 </tr>
 <tr>
 <td>korean_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-korean_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+korean_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/korean_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>60.21</td>
-<td>5.40 / 5.40</td>
+<td>5.40 / 0.97</td>
 <td>9.11 / 4.05</td>
 <td>8.6 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量韩文识别模型，支持韩文、数字识别</td>
 </tr>
 <tr>
 <td>japan_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-japan_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+japan_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/japan_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>45.69</td>
-<td>5.70 / 5.70</td>
+<td>5.70 / 1.02</td>
 <td>8.48 / 4.07</td>
 <td>8.8 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量日文识别模型，支持日文、数字识别</td>
 </tr>
 <tr>
 <td>chinese_cht_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-chinese_cht_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+chinese_cht_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/chinese_cht_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>82.06</td>
-<td>5.90 / 5.90</td>
+<td>5.90 / 1.28</td>
 <td>9.28 / 4.34</td>
 <td>9.7 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量繁体中文识别模型，支持繁体中文、数字识别</td>
 </tr>
 <tr>
 <td>te_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-te_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+te_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/te_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>95.88</td>
-<td>5.42 / 5.42</td>
+<td>5.42 / 0.82</td>
 <td>8.10 / 6.91</td>
 <td>7.8 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量泰卢固文识别模型，支持泰卢固文、数字识别</td>
 </tr>
 <tr>
 <td>ka_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-ka_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+ka_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ka_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>96.96</td>
-<td>5.25 / 5.25</td>
+<td>5.25 / 0.79</td>
 <td>9.09 / 3.86</td>
 <td>8.0 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量卡纳达文识别模型，支持卡纳达文、数字识别</td>
 </tr>
 <tr>
 <td>ta_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-ta_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+ta_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ta_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>76.83</td>
-<td>5.23 / 5.23</td>
+<td>5.23 / 0.75</td>
 <td>10.13 / 4.30</td>
 <td>8.0 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量泰米尔文识别模型，支持泰米尔文、数字识别</td>
 </tr>
 <tr>
 <td>latin_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-latin_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+latin_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/latin_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>76.93</td>
-<td>5.20 / 5.20</td>
+<td>5.20 / 0.79</td>
 <td>8.83 / 7.15</td>
 <td>7.8 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量拉丁文识别模型，支持拉丁文、数字识别</td>
 </tr>
 <tr>
 <td>arabic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-arabic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+arabic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/arabic_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>73.55</td>
-<td>5.35 / 5.35</td>
+<td>5.35 / 0.79</td>
 <td>8.80 / 4.56</td>
 <td>7.8 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量阿拉伯字母识别模型，支持阿拉伯字母、数字识别</td>
 </tr>
 <tr>
 <td>cyrillic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-cyrillic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+cyrillic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/cyrillic_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>94.28</td>
-<td>5.23 / 5.23</td>
+<td>5.23 / 0.76</td>
 <td>8.89 / 3.88</td>
 <td>7.9 M  </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量斯拉夫字母识别模型，支持斯拉夫字母、数字识别</td>
 </tr>
 <tr>
 <td>devanagari_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-devanagari_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+devanagari_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/devanagari_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>96.44</td>
-<td>5.22 / 5.22</td>
+<td>5.22 / 0.79</td>
 <td>8.56 / 4.06</td>
 <td>7.9 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量梵文字母识别模型，支持梵文字母、数字识别</td>

docs/pipeline_usage/tutorials/ocr_pipelines/seal_recognition.en.md

@@ -186,7 +186,7 @@ The seal text recognition pipeline is used to recognize the text content of seal
 </tr>
 </tbody>
 </table>
-
+</details>
 
 <p><b>Document Image Orientation Classification Module (Optional):</b></p>
 <table>
@@ -264,6 +264,7 @@ The seal text recognition pipeline is used to recognize the text content of seal
 </table>
 
 <p><b>Text Recognition Module:</b></p>
+
 <table>
 <tr>
 <th>Model</th><th>Model Download Link</th>
@@ -271,31 +272,88 @@ The seal text recognition pipeline is used to recognize the text content of seal
 <th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
-<th>Description</th>
+<th>Introduction</th>
+</tr>
+<tr>
+<td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_doc_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>81.53</td>
+<td>6.65 / 2.38</td>
+<td>32.92 / 32.92</td>
+<td>74.7 M</td>
+<td>PP-OCRv4_server_rec_doc is trained on a mixed dataset of more Chinese document data and PP-OCR training data based on PP-OCRv4_server_rec. It has added the ability to recognize some traditional Chinese characters, Japanese, and special characters, and can support the recognition of more than 15,000 characters. In addition to improving the text recognition capability related to documents, it also enhances the general text recognition capability.</td>
+</tr>
+<tr>
+<td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Training Model</a></td>
+<td>78.74</td>
+<td>4.82 / 1.20</td>
+<td>16.74 / 4.64</td>
+<td>10.6 M</td>
+<td>
+The lightweight recognition model of PP-OCRv4 has high inference efficiency and can be deployed on various hardware devices, including edge devices.</td>
+</tr>
+<tr>
+<td>PP-OCRv4_server_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Training Model</a></td>
+<td>80.61 </td>
+<td>6.58 / 2.43</td>
+<td>33.17 / 33.17</td>
+<td>71.2 M</td>
+<td>The server-side model of PP-OCRv4 offers high inference accuracy and can be deployed on various types of servers.</td>
+</tr>
+<tr>
+<td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>70.39</td>
+<td>4.81 / 0.75</td>
+<td>16.10 / 5.31</td>
+<td>6.8 M</td>
+<td>The ultra-lightweight English recognition model, trained based on the PP-OCRv4 recognition model, supports the recognition of English letters and numbers.</td>
+</tr>
+</table>
+
+> ❗ The above list features the <b>4 core models</b> that the text recognition module primarily supports. In total, this module supports <b>18 models</b>. The complete list of models is as follows:
+
+<details><summary> 👉Model List Details</summary>
+
+* <b>Chinese Recognition Model</b>
+<table>
+<tr>
+<th>Model</th><th>Model Download Link</th>
+<th>Recognition Avg Accuracy(%)</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>Model Storage Size (M)</th>
+<th>Introduction</th>
 </tr>
 <tr>
-<td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Trained Model</a></td>
-<td>78.20</td>
-<td>4.82 / 4.82</td>
+<td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_doc_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>81.53</td>
+<td>6.65 / 2.38</td>
+<td>32.92 / 32.92</td>
+<td>74.7 M</td>
+<td>PP-OCRv4_server_rec_doc is trained on a mixed dataset of more Chinese document data and PP-OCR training data based on PP-OCRv4_server_rec. It has added the recognition capabilities for some traditional Chinese characters, Japanese, and special characters. The number of recognizable characters is over 15,000. In addition to the improvement in document-related text recognition, it also enhances the general text recognition capability.</td>
+</tr>
+<tr>
+<td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Training Model</a></td>
+<td>78.74</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
-<td>The PP-OCRv4 recognition model is an upgrade from PP-OCRv3. Under comparable speed conditions, the effect in Chinese and English scenarios is further improved. The average recognition accuracy of the 80 multilingual models is increased by more than 8%.</td>
+<td>The lightweight recognition model of PP-OCRv4 has high inference efficiency and can be deployed on various hardware devices, including edge devices.</td>
 </tr>
 <tr>
-<td>PP-OCRv4_server_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Trained Model</a></td>
-<td>79.20</td>
-<td>6.58 / 6.58</td>
+<td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Trained Model</a></td>
+<td>80.61 </td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
-<td>A high-precision server text recognition model, featuring high accuracy, fast speed, and multilingual support. It is suitable for text recognition tasks in various scenarios.</td>
+<td>The server-side model of PP-OCRv4 offers high inference accuracy and can be deployed on various types of servers.</td>
 </tr>
 <tr>
 <td>PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td>5.87 / 5.87</td>
+<td>72.96</td>
+<td>5.87 / 1.19</td>
 <td>9.07 / 4.28</td>
-<td></td>
-<td>An ultra-lightweight OCR model suitable for mobile applications. It adopts an encoder-decoder structure based on Transformer and enhances recognition accuracy and efficiency through techniques such as data augmentation and mixed precision training. The model size is 10.6M, making it suitable for deployment on resource-constrained devices. It can be used in scenarios such as mobile photo translation and business card recognition.</td>
+<td>9.2 M</td>
+<td>PP-OCRv3’s lightweight recognition model is designed for high inference efficiency and can be deployed on a variety of hardware devices, including edge devices.</td>
 </tr>
 </table>
 
@@ -303,16 +361,16 @@ The seal text recognition pipeline is used to recognize the text content of seal
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Recognition Avg Accuracy(%)</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <tr>
 <td>ch_SVTRv2_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_SVTRv2_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_SVTRv2_rec_pretrained.pdparams">Training Model</a></td>
 <td>68.81</td>
-<td>8.36801</td>
-<td>165.706</td>
+<td>8.08 / 2.74</td>
+<td>50.17 / 42.50</td>
 <td>73.9 M</td>
 <td rowspan="1">
 SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan University's Visual and Learning Laboratory (FVL). It won the first prize in the PaddleOCR Algorithm Model Challenge - Task One: OCR End-to-End Recognition Task. The end-to-end recognition accuracy on the A list is 6% higher than that of PP-OCRv4.
@@ -324,16 +382,16 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Recognition Avg Accuracy(%)</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <tr>
 <td>ch_RepSVTR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_RepSVTR_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_RepSVTR_rec_pretrained.pdparams">Training Model</a></td>
 <td>65.07</td>
-<td>10.5047</td>
-<td>51.5647</td>
+<td>5.93 / 1.62</td>
+<td>20.73 / 7.32</td>
 <td>22.1 M</td>
 <td rowspan="1">    The RepSVTR text recognition model is a mobile text recognition model based on SVTRv2. It won the first prize in the PaddleOCR Algorithm Model Challenge - Task One: OCR End-to-End Recognition Task. The end-to-end recognition accuracy on the B list is 2.5% higher than that of PP-OCRv4, with the same inference speed.</td>
 </tr>
@@ -344,26 +402,26 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Recognition Avg Accuracy(%)</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <tr>
 <td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>[Latest] Further upgraded based on PP-OCRv3, with improved accuracy under comparable speed conditions.</td>
+<td> 70.39</td>
+<td>4.81 / 0.75</td>
+<td>16.10 / 5.31</td>
+<td>6.8 M</td>
+<td>The ultra-lightweight English recognition model trained based on the PP-OCRv4 recognition model supports the recognition of English and numbers.</td>
 </tr>
 <tr>
 <td>en_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Ultra-lightweight model, supporting English and numeric recognition.</td>
+<td>70.69</td>
+<td>5.44 / 0.75</td>
+<td>8.65 / 5.57</td>
+<td>7.8 M </td>
+<td>The ultra-lightweight English recognition model trained based on the PP-OCRv3 recognition model supports the recognition of English and numbers.</td>
 </tr>
 </table>
 
@@ -372,92 +430,94 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Recognition Avg Accuracy(%)</th>
-<th>GPU Inference Time (ms)</th>
-<th>CPU Inference Time</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <tr>
 <td>korean_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/korean_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Korean Recognition</td>
+<td>60.21</td>
+<td>5.40 / 0.97</td>
+<td>9.11 / 4.05</td>
+<td>8.6 M</td>
+<td>The ultra-lightweight Korean recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Korean and numbers. </td>
 </tr>
 <tr>
 <td>japan_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/japan_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Japanese Recognition</td>
+<td>45.69</td>
+<td>5.70 / 1.02</td>
+<td>8.48 / 4.07</td>
+<td>8.8 M </td>
+<td>The ultra-lightweight Japanese recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Japanese and numbers.</td>
 </tr>
 <tr>
 <td>chinese_cht_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/chinese_cht_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Traditional Chinese Recognition</td>
+<td>82.06</td>
+<td>5.90 / 1.28</td>
+<td>9.28 / 4.34</td>
+<td>9.7 M </td>
+<td>The ultra-lightweight Traditional Chinese recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Traditional Chinese and numbers.</td>
 </tr>
 <tr>
 <td>te_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/te_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Telugu Recognition</td>
+<td>95.88</td>
+<td>5.42 / 0.82</td>
+<td>8.10 / 6.91</td>
+<td>7.8 M </td>
+<td>The ultra-lightweight Telugu recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Telugu and numbers.</td>
 </tr>
 <tr>
 <td>ka_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ka_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Kannada Recognition</td>
+<td>96.96</td>
+<td>5.25 / 0.79</td>
+<td>9.09 / 3.86</td>
+<td>8.0 M </td>
+<td>The ultra-lightweight Kannada recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Kannada and numbers.</td>
 </tr>
 <tr>
 <td>ta_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ta_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Tamil Recognition</td>
+<td>76.83</td>
+<td>5.23 / 0.75</td>
+<td>10.13 / 4.30</td>
+<td>8.0 M </td>
+<td>The ultra-lightweight Tamil recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Tamil and numbers.</td>
 </tr>
 <tr>
 <td>latin_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/latin_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Latin Recognition</td>
+<td>76.93</td>
+<td>5.20 / 0.79</td>
+<td>8.83 / 7.15</td>
+<td>7.8 M</td>
+<td>The ultra-lightweight Latin recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Latin script and numbers.</td>
 </tr>
 <tr>
 <td>arabic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/arabic_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Arabic Script Recognition</td>
+<td>73.55</td>
+<td>5.35 / 0.79</td>
+<td>8.80 / 4.56</td>
+<td>7.8 M</td>
+<td>The ultra-lightweight Arabic script recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Arabic script and numbers.</td>
 </tr>
 <tr>
 <td>cyrillic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/cyrillic_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Cyrillic Script Recognition</td>
+<td>94.28</td>
+<td>5.23 / 0.76</td>
+<td>8.89 / 3.88</td>
+<td>7.9 M  </td>
+<td>
+The ultra-lightweight cyrillic alphabet recognition model trained based on the PP-OCRv3 recognition model supports the recognition of cyrillic letters and numbers.</td>
 </tr>
 <tr>
 <td>devanagari_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/devanagari_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
-<td></td>
-<td></td>
-<td></td>
-<td></td>
-<td>Devanagari Script Recognition</td>
+<td>96.44</td>
+<td>5.22 / 0.79</td>
+<td>8.56 / 4.06</td>
+<td>7.9 M  </td>
+<td>The ultra-lightweight Devanagari script recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Devanagari script and numbers.</td>
 </tr>
 </table>
+</details>
 
 **Test Environment Description**:
 
@@ -488,7 +548,7 @@ SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan
 | Normal Mode | FP32 Precision / No TRT Acceleration   | FP32 Precision / 8 Threads | PaddleInference                                 |
 | High-Performance Mode | Optimal combination of pre-selected precision types and acceleration strategies | FP32 Precision / 8 Threads | Pre-selected optimal backend (Paddle/OpenVINO/TRT, etc.) |
 
-</details>
+
 
 ## 2. Quick Start
 All model pipelines provided by PaddleX can be quickly experienced. You can experience the effect of the seal text recognition pipeline on the community platform, or you can use the command line or Python locally to experience the effect of the seal text recognition pipeline.

docs/pipeline_usage/tutorials/ocr_pipelines/seal_recognition.md

@@ -231,9 +231,9 @@ comments: true
 </tr>
 <tr>
 <td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_doc_pretrained.pdparams">训练模型</a></td>
 <td>81.53</td>
-<td>6.65 / 6.65</td>
+<td>6.65 / 2.38</td>
 <td>32.92 / 32.92</td>
 <td>74.7 M</td>
 <td>PP-OCRv4_server_rec_doc是在PP-OCRv4_server_rec的基础上，在更多中文文档数据和PP-OCR训练数据的混合数据训练而成，增加了部分繁体字、日文、特殊字符的识别能力，可支持识别的字符为1.5万+，除文档相关的文字识别能力提升外，也同时提升了通用文字的识别能力</td>
@@ -241,7 +241,7 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a><
 <tr>
 <td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>78.74</td>
-<td>4.82 / 4.82</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
 <td>PP-OCRv4的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
@@ -249,23 +249,23 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a><
 <tr>
 <td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">训练模型</a></td>
 <td>80.61 </td>
-<td>6.58 / 6.58</td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
 <td>PP-OCRv4的服务器端模型，推理精度高，可以部署在多种不同的服务器上</td>
 </tr>
 <tr>
 <td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>70.39</td>
-<td>4.81 / 4.81</td>
+<td>4.81 / 0.75</td>
 <td>16.10 / 5.31</td>
 <td>6.8 M</td>
 <td>基于PP-OCRv4识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
 </tr>
 </table>
 
->❗ 以上列出的是文本识别模块重点支持的<b>4个核心模型</b>，该模块总共支持<b>18个全量模型</b>，包含多个多语言文本识别模型，完整的模型列表如下：
+> ❗ 以上列出的是文本识别模块重点支持的<b>4个核心模型</b>，该模块总共支持<b>18个全量模型</b>，包含多个多语言文本识别模型，完整的模型列表如下：
 
 <details><summary> 👉模型列表详情</summary>
 
@@ -281,9 +281,9 @@ en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></
 </tr>
 <tr>
 <td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_doc_pretrained.pdparams">训练模型</a></td>
 <td>81.53</td>
-<td>6.65 / 6.65</td>
+<td>6.65 / 2.38</td>
 <td>32.92 / 32.92</td>
 <td>74.7 M</td>
 <td>PP-OCRv4_server_rec_doc是在PP-OCRv4_server_rec的基础上，在更多中文文档数据和PP-OCR训练数据的混合数据训练而成，增加了部分繁体字、日文、特殊字符的识别能力，可支持识别的字符为1.5万+，除文档相关的文字识别能力提升外，也同时提升了通用文字的识别能力</td>
@@ -291,7 +291,7 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a><
 <tr>
 <td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>78.74</td>
-<td>4.82 / 4.82</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
 <td>PP-OCRv4的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
@@ -299,16 +299,16 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a><
 <tr>
 <td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">训练模型</a></td>
 <td>80.61 </td>
-<td>6.58 / 6.58</td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
 <td>PP-OCRv4的服务器端模型，推理精度高，可以部署在多种不同的服务器上</td>
 </tr>
 <tr>
 <td>PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>72.96</td>
-<td>5.87 / 5.87</td>
+<td>5.87 / 1.19</td>
 <td>9.07 / 4.28</td>
 <td>9.2 M</td>
 <td>PP-OCRv3的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
@@ -327,7 +327,7 @@ PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
 <tr>
 <td>ch_SVTRv2_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_SVTRv2_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_SVTRv2_rec_pretrained.pdparams">训练模型</a></td>
 <td>68.81</td>
-<td>8.08 / 8.08</td>
+<td>8.08 / 2.74</td>
 <td>50.17 / 42.50</td>
 <td>73.9 M</td>
 <td rowspan="1">
@@ -348,7 +348,7 @@ SVTRv2 是一种由复旦大学视觉与学习实验室（FVL）的OpenOCR团队
 <tr>
 <td>ch_RepSVTR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_RepSVTR_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_RepSVTR_rec_pretrained.pdparams">训练模型</a></td>
 <td>65.07</td>
-<td>5.93 / 5.93</td>
+<td>5.93 / 1.62</td>
 <td>20.73 / 7.32</td>
 <td>22.1 M</td>
 <td rowspan="1">    RepSVTR 文本识别模型是一种基于SVTRv2 的移动端文本识别模型，其在PaddleOCR算法模型挑战赛 - 赛题一：OCR端到端识别任务中荣获一等奖，B榜端到端识别精度相比PP-OCRv4提升2.5%，推理速度持平。</td>
@@ -367,18 +367,18 @@ SVTRv2 是一种由复旦大学视觉与学习实验室（FVL）的OpenOCR团队
 </tr>
 <tr>
 <td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td> 70.39</td>
-<td>4.81 / 4.81</td>
+<td>4.81 / 0.75</td>
 <td>16.10 / 5.31</td>
 <td>6.8 M</td>
 <td>基于PP-OCRv4识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
 </tr>
 <tr>
 <td>en_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-en_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+en_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>70.69</td>
-<td>5.44 / 5.44</td>
+<td>5.44 / 0.75</td>
 <td>8.65 / 5.57</td>
 <td>7.8 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
@@ -397,95 +397,96 @@ en_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></
 </tr>
 <tr>
 <td>korean_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-korean_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+korean_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/korean_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>60.21</td>
-<td>5.40 / 5.40</td>
+<td>5.40 / 0.97</td>
 <td>9.11 / 4.05</td>
 <td>8.6 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量韩文识别模型，支持韩文、数字识别</td>
 </tr>
 <tr>
 <td>japan_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-japan_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+japan_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/japan_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>45.69</td>
-<td>5.70 / 5.70</td>
+<td>5.70 / 1.02</td>
 <td>8.48 / 4.07</td>
 <td>8.8 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量日文识别模型，支持日文、数字识别</td>
 </tr>
 <tr>
 <td>chinese_cht_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-chinese_cht_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+chinese_cht_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/chinese_cht_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>82.06</td>
-<td>5.90 / 5.90</td>
+<td>5.90 / 1.28</td>
 <td>9.28 / 4.34</td>
 <td>9.7 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量繁体中文识别模型，支持繁体中文、数字识别</td>
 </tr>
 <tr>
 <td>te_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-te_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+te_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/te_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>95.88</td>
-<td>5.42 / 5.42</td>
+<td>5.42 / 0.82</td>
 <td>8.10 / 6.91</td>
 <td>7.8 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量泰卢固文识别模型，支持泰卢固文、数字识别</td>
 </tr>
 <tr>
 <td>ka_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-ka_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+ka_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ka_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>96.96</td>
-<td>5.25 / 5.25</td>
+<td>5.25 / 0.79</td>
 <td>9.09 / 3.86</td>
 <td>8.0 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量卡纳达文识别模型，支持卡纳达文、数字识别</td>
 </tr>
 <tr>
 <td>ta_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-ta_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+ta_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ta_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>76.83</td>
-<td>5.23 / 5.23</td>
+<td>5.23 / 0.75</td>
 <td>10.13 / 4.30</td>
 <td>8.0 M </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量泰米尔文识别模型，支持泰米尔文、数字识别</td>
 </tr>
 <tr>
 <td>latin_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-latin_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+latin_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/latin_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>76.93</td>
-<td>5.20 / 5.20</td>
+<td>5.20 / 0.79</td>
 <td>8.83 / 7.15</td>
 <td>7.8 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量拉丁文识别模型，支持拉丁文、数字识别</td>
 </tr>
 <tr>
 <td>arabic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-arabic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+arabic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/arabic_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>73.55</td>
-<td>5.35 / 5.35</td>
+<td>5.35 / 0.79</td>
 <td>8.80 / 4.56</td>
 <td>7.8 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量阿拉伯字母识别模型，支持阿拉伯字母、数字识别</td>
 </tr>
 <tr>
 <td>cyrillic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-cyrillic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+cyrillic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/cyrillic_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>94.28</td>
-<td>5.23 / 5.23</td>
+<td>5.23 / 0.76</td>
 <td>8.89 / 3.88</td>
 <td>7.9 M  </td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量斯拉夫字母识别模型，支持斯拉夫字母、数字识别</td>
 </tr>
 <tr>
 <td>devanagari_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
-devanagari_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
+devanagari_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/devanagari_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
 <td>96.44</td>
-<td>5.22 / 5.22</td>
+<td>5.22 / 0.79</td>
 <td>8.56 / 4.06</td>
 <td>7.9 M</td>
 <td>基于PP-OCRv3识别模型训练得到的超轻量梵文字母识别模型，支持梵文字母、数字识别</td>
 </tr>
 </table>
+</details>
 
 **测试环境说明：**
 
@@ -516,7 +517,6 @@ devanagari_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模
 | 常规模式    | FP32精度 / 无TRT加速             | FP32精度 / 8线程       | PaddleInference                             |
 | 高性能模式  | 选择先验精度类型和加速策略的最优组合         | FP32精度 / 8线程       | 选择先验最优后端（Paddle/OpenVINO/TRT等） |
 
-</details>
 
 
 ## 2. 快速开始

docs/pipeline_usage/tutorials/ocr_pipelines/table_recognition.en.md

@@ -150,6 +150,261 @@ SLANet_plus is an enhanced version of SLANet, a table structure recognition mode
 </tbody>
 </table>
 
+<p><b>Text Recognition Module Models:</b></p>
+<table>
+<tr>
+<th>Model</th><th>Model Download Link</th>
+<th>Recognition Avg Accuracy(%)</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>Model Storage Size (M)</th>
+<th>Introduction</th>
+</tr>
+<tr>
+<td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_doc_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>81.53</td>
+<td>6.65 / 2.38</td>
+<td>32.92 / 32.92</td>
+<td>74.7 M</td>
+<td>PP-OCRv4_server_rec_doc is trained on a mixed dataset of more Chinese document data and PP-OCR training data based on PP-OCRv4_server_rec. It has added the ability to recognize some traditional Chinese characters, Japanese, and special characters, and can support the recognition of more than 15,000 characters. In addition to improving the text recognition capability related to documents, it also enhances the general text recognition capability.</td>
+</tr>
+<tr>
+<td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Training Model</a></td>
+<td>78.74</td>
+<td>4.82 / 1.20</td>
+<td>16.74 / 4.64</td>
+<td>10.6 M</td>
+<td>
+The lightweight recognition model of PP-OCRv4 has high inference efficiency and can be deployed on various hardware devices, including edge devices.</td>
+</tr>
+<tr>
+<td>PP-OCRv4_server_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Training Model</a></td>
+<td>80.61 </td>
+<td>6.58 / 2.43</td>
+<td>33.17 / 33.17</td>
+<td>71.2 M</td>
+<td>The server-side model of PP-OCRv4 offers high inference accuracy and can be deployed on various types of servers.</td>
+</tr>
+<tr>
+<td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>70.39</td>
+<td>4.81 / 0.75</td>
+<td>16.10 / 5.31</td>
+<td>6.8 M</td>
+<td>The ultra-lightweight English recognition model, trained based on the PP-OCRv4 recognition model, supports the recognition of English letters and numbers.</td>
+</tr>
+</table>
+
+> ❗ The above list features the <b>4 core models</b> that the text recognition module primarily supports. In total, this module supports <b>18 models</b>. The complete list of models is as follows:
+
+<details><summary> 👉Model List Details</summary>
+
+* <b>Chinese Recognition Model</b>
+<table>
+<tr>
+<th>Model</th><th>Model Download Link</th>
+<th>Recognition Avg Accuracy(%)</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>Model Storage Size (M)</th>
+<th>Introduction</th>
+</tr>
+<tr>
+<td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_doc_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>81.53</td>
+<td>6.65 / 2.38</td>
+<td>32.92 / 32.92</td>
+<td>74.7 M</td>
+<td>PP-OCRv4_server_rec_doc is trained on a mixed dataset of more Chinese document data and PP-OCR training data based on PP-OCRv4_server_rec. It has added the recognition capabilities for some traditional Chinese characters, Japanese, and special characters. The number of recognizable characters is over 15,000. In addition to the improvement in document-related text recognition, it also enhances the general text recognition capability.</td>
+</tr>
+<tr>
+<td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Training Model</a></td>
+<td>78.74</td>
+<td>4.82 / 1.20</td>
+<td>16.74 / 4.64</td>
+<td>10.6 M</td>
+<td>The lightweight recognition model of PP-OCRv4 has high inference efficiency and can be deployed on various hardware devices, including edge devices.</td>
+</tr>
+<tr>
+<td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Trained Model</a></td>
+<td>80.61 </td>
+<td>6.58 / 2.43</td>
+<td>33.17 / 33.17</td>
+<td>71.2 M</td>
+<td>The server-side model of PP-OCRv4 offers high inference accuracy and can be deployed on various types of servers.</td>
+</tr>
+<tr>
+<td>PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>72.96</td>
+<td>5.87 / 1.19</td>
+<td>9.07 / 4.28</td>
+<td>9.2 M</td>
+<td>PP-OCRv3’s lightweight recognition model is designed for high inference efficiency and can be deployed on a variety of hardware devices, including edge devices.</td>
+</tr>
+</table>
+
+<table>
+<tr>
+<th>Model</th><th>Model Download Link</th>
+<th>Recognition Avg Accuracy(%)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>Model Storage Size (M)</th>
+<th>Introduction</th>
+</tr>
+<tr>
+<td>ch_SVTRv2_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_SVTRv2_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_SVTRv2_rec_pretrained.pdparams">Training Model</a></td>
+<td>68.81</td>
+<td>8.08 / 2.74</td>
+<td>50.17 / 42.50</td>
+<td>73.9 M</td>
+<td rowspan="1">
+SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan University's Visual and Learning Laboratory (FVL). It won the first prize in the PaddleOCR Algorithm Model Challenge - Task One: OCR End-to-End Recognition Task. The end-to-end recognition accuracy on the A list is 6% higher than that of PP-OCRv4.
+</td>
+</tr>
+</table>
+
+<table>
+<tr>
+<th>Model</th><th>Model Download Link</th>
+<th>Recognition Avg Accuracy(%)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>Model Storage Size (M)</th>
+<th>Introduction</th>
+</tr>
+<tr>
+<td>ch_RepSVTR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_RepSVTR_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_RepSVTR_rec_pretrained.pdparams">Training Model</a></td>
+<td>65.07</td>
+<td>5.93 / 1.62</td>
+<td>20.73 / 7.32</td>
+<td>22.1 M</td>
+<td rowspan="1">    The RepSVTR text recognition model is a mobile text recognition model based on SVTRv2. It won the first prize in the PaddleOCR Algorithm Model Challenge - Task One: OCR End-to-End Recognition Task. The end-to-end recognition accuracy on the B list is 2.5% higher than that of PP-OCRv4, with the same inference speed.</td>
+</tr>
+</table>
+
+* <b>English Recognition Model</b>
+<table>
+<tr>
+<th>Model</th><th>Model Download Link</th>
+<th>Recognition Avg Accuracy(%)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>Model Storage Size (M)</th>
+<th>Introduction</th>
+</tr>
+<tr>
+<td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td> 70.39</td>
+<td>4.81 / 0.75</td>
+<td>16.10 / 5.31</td>
+<td>6.8 M</td>
+<td>The ultra-lightweight English recognition model trained based on the PP-OCRv4 recognition model supports the recognition of English and numbers.</td>
+</tr>
+<tr>
+<td>en_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>70.69</td>
+<td>5.44 / 0.75</td>
+<td>8.65 / 5.57</td>
+<td>7.8 M </td>
+<td>The ultra-lightweight English recognition model trained based on the PP-OCRv3 recognition model supports the recognition of English and numbers.</td>
+</tr>
+</table>
+
+* <b>Multilingual Recognition Model</b>
+<table>
+<tr>
+<th>Model</th><th>Model Download Link</th>
+<th>Recognition Avg Accuracy(%)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>Model Storage Size (M)</th>
+<th>Introduction</th>
+</tr>
+<tr>
+<td>korean_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/korean_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>60.21</td>
+<td>5.40 / 0.97</td>
+<td>9.11 / 4.05</td>
+<td>8.6 M</td>
+<td>The ultra-lightweight Korean recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Korean and numbers. </td>
+</tr>
+<tr>
+<td>japan_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/japan_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>45.69</td>
+<td>5.70 / 1.02</td>
+<td>8.48 / 4.07</td>
+<td>8.8 M </td>
+<td>The ultra-lightweight Japanese recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Japanese and numbers.</td>
+</tr>
+<tr>
+<td>chinese_cht_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/chinese_cht_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>82.06</td>
+<td>5.90 / 1.28</td>
+<td>9.28 / 4.34</td>
+<td>9.7 M </td>
+<td>The ultra-lightweight Traditional Chinese recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Traditional Chinese and numbers.</td>
+</tr>
+<tr>
+<td>te_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/te_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>95.88</td>
+<td>5.42 / 0.82</td>
+<td>8.10 / 6.91</td>
+<td>7.8 M </td>
+<td>The ultra-lightweight Telugu recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Telugu and numbers.</td>
+</tr>
+<tr>
+<td>ka_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ka_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>96.96</td>
+<td>5.25 / 0.79</td>
+<td>9.09 / 3.86</td>
+<td>8.0 M </td>
+<td>The ultra-lightweight Kannada recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Kannada and numbers.</td>
+</tr>
+<tr>
+<td>ta_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ta_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>76.83</td>
+<td>5.23 / 0.75</td>
+<td>10.13 / 4.30</td>
+<td>8.0 M </td>
+<td>The ultra-lightweight Tamil recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Tamil and numbers.</td>
+</tr>
+<tr>
+<td>latin_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/latin_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>76.93</td>
+<td>5.20 / 0.79</td>
+<td>8.83 / 7.15</td>
+<td>7.8 M</td>
+<td>The ultra-lightweight Latin recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Latin script and numbers.</td>
+</tr>
+<tr>
+<td>arabic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/arabic_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>73.55</td>
+<td>5.35 / 0.79</td>
+<td>8.80 / 4.56</td>
+<td>7.8 M</td>
+<td>The ultra-lightweight Arabic script recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Arabic script and numbers.</td>
+</tr>
+<tr>
+<td>cyrillic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/cyrillic_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>94.28</td>
+<td>5.23 / 0.76</td>
+<td>8.89 / 3.88</td>
+<td>7.9 M  </td>
+<td>
+The ultra-lightweight cyrillic alphabet recognition model trained based on the PP-OCRv3 recognition model supports the recognition of cyrillic letters and numbers.</td>
+</tr>
+<tr>
+<td>devanagari_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/devanagari_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>96.44</td>
+<td>5.22 / 0.79</td>
+<td>8.56 / 4.06</td>
+<td>7.9 M  </td>
+<td>The ultra-lightweight Devanagari script recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Devanagari script and numbers.</td>
+</tr>
+</table>
+</details>
+
 **Test Environment Description**:
 
 - **Performance Test Environment**

docs/pipeline_usage/tutorials/ocr_pipelines/table_recognition.md

@@ -86,19 +86,88 @@ comments: true
 <th>介绍</th>
 </tr>
 <tr>
+<td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_doc_pretrained.pdparams">训练模型</a></td>
+<td>81.53</td>
+<td>6.65 / 2.38</td>
+<td>32.92 / 32.92</td>
+<td>74.7 M</td>
+<td>PP-OCRv4_server_rec_doc是在PP-OCRv4_server_rec的基础上，在更多中文文档数据和PP-OCR训练数据的混合数据训练而成，增加了部分繁体字、日文、特殊字符的识别能力，可支持识别的字符为1.5万+，除文档相关的文字识别能力提升外，也同时提升了通用文字的识别能力</td>
+</tr>
+<tr>
 <td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
-<td>78.20</td>
-<td>4.82 / 4.82</td>
+<td>78.74</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
-<td rowspan="2">PP-OCRv4是百度飞桨视觉团队自研的文本识别模型PP-OCRv3的下一个版本，通过引入数据增强方案、GTC-NRTR指导分支等策略，在模型推理速度不变的情况下，进一步提升了文本识别精度。该模型提供了服务端（server）和移动端（mobile）两个不同版本，来满足不同场景下的工业需求。</td>
+<td>PP-OCRv4的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
 </tr>
 <tr>
 <td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">训练模型</a></td>
-<td>79.20</td>
-<td>6.58 / 6.58</td>
+<td>80.61 </td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
+<td>PP-OCRv4的服务器端模型，推理精度高，可以部署在多种不同的服务器上</td>
+</tr>
+<tr>
+<td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>70.39</td>
+<td>4.81 / 0.75</td>
+<td>16.10 / 5.31</td>
+<td>6.8 M</td>
+<td>基于PP-OCRv4识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
+</tr>
+</table>
+
+> ❗ 以上列出的是文本识别模块重点支持的<b>4个核心模型</b>，该模块总共支持<b>18个全量模型</b>，包含多个多语言文本识别模型，完整的模型列表如下：
+
+<details><summary> 👉模型列表详情</summary>
+
+* <b>中文识别模型</b>
+<table>
+<tr>
+<th>模型</th><th>模型下载链接</th>
+<th>识别 Avg Accuracy(%)</th>
+<th>GPU推理耗时（ms）<br/>[常规模式 / 高性能模式]</th>
+<th>CPU推理耗时（ms）<br/>[常规模式 / 高性能模式]</th>
+<th>模型存储大小（M）</th>
+<th>介绍</th>
+</tr>
+<tr>
+<td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_doc_pretrained.pdparams">训练模型</a></td>
+<td>81.53</td>
+<td>6.65 / 2.38</td>
+<td>32.92 / 32.92</td>
+<td>74.7 M</td>
+<td>PP-OCRv4_server_rec_doc是在PP-OCRv4_server_rec的基础上，在更多中文文档数据和PP-OCR训练数据的混合数据训练而成，增加了部分繁体字、日文、特殊字符的识别能力，可支持识别的字符为1.5万+，除文档相关的文字识别能力提升外，也同时提升了通用文字的识别能力</td>
+</tr>
+<tr>
+<td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>78.74</td>
+<td>4.82 / 1.20</td>
+<td>16.74 / 4.64</td>
+<td>10.6 M</td>
+<td>PP-OCRv4的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
+</tr>
+<tr>
+<td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">训练模型</a></td>
+<td>80.61 </td>
+<td>6.58 / 2.43</td>
+<td>33.17 / 33.17</td>
+<td>71.2 M</td>
+<td>PP-OCRv4的服务器端模型，推理精度高，可以部署在多种不同的服务器上</td>
+</tr>
+<tr>
+<td>PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>72.96</td>
+<td>5.87 / 1.19</td>
+<td>9.07 / 4.28</td>
+<td>9.2 M</td>
+<td>PP-OCRv3的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
 </tr>
 </table>
 
@@ -114,7 +183,7 @@ comments: true
 <tr>
 <td>ch_SVTRv2_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_SVTRv2_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_SVTRv2_rec_pretrained.pdparams">训练模型</a></td>
 <td>68.81</td>
-<td>8.08 / 8.08</td>
+<td>8.08 / 2.74</td>
 <td>50.17 / 42.50</td>
 <td>73.9 M</td>
 <td rowspan="1">
@@ -135,13 +204,147 @@ SVTRv2 是一种由复旦大学视觉与学习实验室（FVL）的OpenOCR团队
 <tr>
 <td>ch_RepSVTR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_RepSVTR_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_RepSVTR_rec_pretrained.pdparams">训练模型</a></td>
 <td>65.07</td>
-<td>5.93 / 5.93</td>
+<td>5.93 / 1.62</td>
 <td>20.73 / 7.32</td>
 <td>22.1 M</td>
 <td rowspan="1">    RepSVTR 文本识别模型是一种基于SVTRv2 的移动端文本识别模型，其在PaddleOCR算法模型挑战赛 - 赛题一：OCR端到端识别任务中荣获一等奖，B榜端到端识别精度相比PP-OCRv4提升2.5%，推理速度持平。</td>
 </tr>
 </table>
 
+* <b>英文识别模型</b>
+<table>
+<tr>
+<th>模型</th><th>模型下载链接</th>
+<th>识别 Avg Accuracy(%)</th>
+<th>GPU推理耗时（ms）<br/>[常规模式 / 高性能模式]</th>
+<th>CPU推理耗时（ms）<br/>[常规模式 / 高性能模式]</th>
+<th>模型存储大小（M）</th>
+<th>介绍</th>
+</tr>
+<tr>
+<td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td> 70.39</td>
+<td>4.81 / 0.75</td>
+<td>16.10 / 5.31</td>
+<td>6.8 M</td>
+<td>基于PP-OCRv4识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
+</tr>
+<tr>
+<td>en_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+en_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>70.69</td>
+<td>5.44 / 0.75</td>
+<td>8.65 / 5.57</td>
+<td>7.8 M </td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
+</tr>
+</table>
+
+
+* <b>多语言识别模型</b>
+<table>
+<tr>
+<th>模型</th><th>模型下载链接</th>
+<th>识别 Avg Accuracy(%)</th>
+<th>GPU推理耗时（ms）<br/>[常规模式 / 高性能模式]</th>
+<th>CPU推理耗时（ms）<br/>[常规模式 / 高性能模式]</th>
+<th>模型存储大小（M）</th>
+<th>介绍</th>
+</tr>
+<tr>
+<td>korean_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+korean_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/korean_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>60.21</td>
+<td>5.40 / 0.97</td>
+<td>9.11 / 4.05</td>
+<td>8.6 M</td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量韩文识别模型，支持韩文、数字识别</td>
+</tr>
+<tr>
+<td>japan_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+japan_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/japan_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>45.69</td>
+<td>5.70 / 1.02</td>
+<td>8.48 / 4.07</td>
+<td>8.8 M </td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量日文识别模型，支持日文、数字识别</td>
+</tr>
+<tr>
+<td>chinese_cht_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+chinese_cht_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/chinese_cht_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>82.06</td>
+<td>5.90 / 1.28</td>
+<td>9.28 / 4.34</td>
+<td>9.7 M </td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量繁体中文识别模型，支持繁体中文、数字识别</td>
+</tr>
+<tr>
+<td>te_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+te_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/te_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>95.88</td>
+<td>5.42 / 0.82</td>
+<td>8.10 / 6.91</td>
+<td>7.8 M </td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量泰卢固文识别模型，支持泰卢固文、数字识别</td>
+</tr>
+<tr>
+<td>ka_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+ka_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ka_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>96.96</td>
+<td>5.25 / 0.79</td>
+<td>9.09 / 3.86</td>
+<td>8.0 M </td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量卡纳达文识别模型，支持卡纳达文、数字识别</td>
+</tr>
+<tr>
+<td>ta_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+ta_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ta_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>76.83</td>
+<td>5.23 / 0.75</td>
+<td>10.13 / 4.30</td>
+<td>8.0 M </td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量泰米尔文识别模型，支持泰米尔文、数字识别</td>
+</tr>
+<tr>
+<td>latin_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+latin_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/latin_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>76.93</td>
+<td>5.20 / 0.79</td>
+<td>8.83 / 7.15</td>
+<td>7.8 M</td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量拉丁文识别模型，支持拉丁文、数字识别</td>
+</tr>
+<tr>
+<td>arabic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+arabic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/arabic_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>73.55</td>
+<td>5.35 / 0.79</td>
+<td>8.80 / 4.56</td>
+<td>7.8 M</td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量阿拉伯字母识别模型，支持阿拉伯字母、数字识别</td>
+</tr>
+<tr>
+<td>cyrillic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+cyrillic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/cyrillic_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>94.28</td>
+<td>5.23 / 0.76</td>
+<td>8.89 / 3.88</td>
+<td>7.9 M  </td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量斯拉夫字母识别模型，支持斯拉夫字母、数字识别</td>
+</tr>
+<tr>
+<td>devanagari_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+devanagari_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/devanagari_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>96.44</td>
+<td>5.22 / 0.79</td>
+<td>8.56 / 4.06</td>
+<td>7.9 M</td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量梵文字母识别模型，支持梵文字母、数字识别</td>
+</tr>
+</table>
+</details>
+
 <p><b>版面区域检测模块模型（可选）：</b></p>
 <table>
 <thead>

docs/pipeline_usage/tutorials/ocr_pipelines/table_recognition_v2.en.md

@@ -135,22 +135,45 @@ The General Table Recognition v2 Pipeline is designed to solve table recognition
 <th>Introduction</th>
 </tr>
 <tr>
-<td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Training Model</a></td>
-<td>78.20</td>
-<td>4.82 / 4.82</td>
+<td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_doc_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>81.53</td>
+<td>6.65 / 2.38</td>
+<td>32.92 / 32.92</td>
+<td>74.7 M</td>
+<td>PP-OCRv4_server_rec_doc is trained on a mixed dataset of more Chinese document data and PP-OCR training data based on PP-OCRv4_server_rec. It has added the ability to recognize some traditional Chinese characters, Japanese, and special characters, and can support the recognition of more than 15,000 characters. In addition to improving the text recognition capability related to documents, it also enhances the general text recognition capability.</td>
+</tr>
+<tr>
+<td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Training Model</a></td>
+<td>78.74</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
-<td rowspan="2">PP-OCRv4 is the next version of the self-developed text recognition model PP-OCRv3 by Baidu PaddlePaddle Vision Team. By introducing data augmentation schemes and GTC-NRTR guidance branches, it further improves text recognition accuracy without changing the model inference speed. This model provides both server and mobile versions to meet industrial needs in different scenarios.</td>
+<td>
+The lightweight recognition model of PP-OCRv4 has high inference efficiency and can be deployed on various hardware devices, including edge devices.</td>
 </tr>
 <tr>
-<td>PP-OCRv4_server_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Training Model</a></td>
-<td>79.20</td>
-<td>6.58 / 6.58</td>
+<td>PP-OCRv4_server_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Training Model</a></td>
+<td>80.61 </td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
+<td>The server-side model of PP-OCRv4 offers high inference accuracy and can be deployed on various types of servers.</td>
+</tr>
+<tr>
+<td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>70.39</td>
+<td>4.81 / 0.75</td>
+<td>16.10 / 5.31</td>
+<td>6.8 M</td>
+<td>The ultra-lightweight English recognition model, trained based on the PP-OCRv4 recognition model, supports the recognition of English letters and numbers.</td>
 </tr>
 </table>
 
+> ❗ The above list features the <b>4 core models</b> that the text recognition module primarily supports. In total, this module supports <b>18 models</b>. The complete list of models is as follows:
+
+<details><summary> 👉Model List Details</summary>
+
+* <b>Chinese Recognition Model</b>
 <table>
 <tr>
 <th>Model</th><th>Model Download Link</th>
@@ -161,13 +184,56 @@ The General Table Recognition v2 Pipeline is designed to solve table recognition
 <th>Introduction</th>
 </tr>
 <tr>
-<td>ch_SVTRv2_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/ch_SVTRv2_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_SVTRv2_rec_pretrained.pdparams">Training Model</a></td>
+<td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_doc_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>81.53</td>
+<td>6.65 / 2.38</td>
+<td>32.92 / 32.92</td>
+<td>74.7 M</td>
+<td>PP-OCRv4_server_rec_doc is trained on a mixed dataset of more Chinese document data and PP-OCR training data based on PP-OCRv4_server_rec. It has added the recognition capabilities for some traditional Chinese characters, Japanese, and special characters. The number of recognizable characters is over 15,000. In addition to the improvement in document-related text recognition, it also enhances the general text recognition capability.</td>
+</tr>
+<tr>
+<td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">Training Model</a></td>
+<td>78.74</td>
+<td>4.82 / 1.20</td>
+<td>16.74 / 4.64</td>
+<td>10.6 M</td>
+<td>The lightweight recognition model of PP-OCRv4 has high inference efficiency and can be deployed on various hardware devices, including edge devices.</td>
+</tr>
+<tr>
+<td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">Trained Model</a></td>
+<td>80.61 </td>
+<td>6.58 / 2.43</td>
+<td>33.17 / 33.17</td>
+<td>71.2 M</td>
+<td>The server-side model of PP-OCRv4 offers high inference accuracy and can be deployed on various types of servers.</td>
+</tr>
+<tr>
+<td>PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>72.96</td>
+<td>5.87 / 1.19</td>
+<td>9.07 / 4.28</td>
+<td>9.2 M</td>
+<td>PP-OCRv3’s lightweight recognition model is designed for high inference efficiency and can be deployed on a variety of hardware devices, including edge devices.</td>
+</tr>
+</table>
+
+<table>
+<tr>
+<th>Model</th><th>Model Download Link</th>
+<th>Recognition Avg Accuracy(%)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>Model Storage Size (M)</th>
+<th>Introduction</th>
+</tr>
+<tr>
+<td>ch_SVTRv2_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_SVTRv2_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_SVTRv2_rec_pretrained.pdparams">Training Model</a></td>
 <td>68.81</td>
-<td>8.08 / 8.08</td>
+<td>8.08 / 2.74</td>
 <td>50.17 / 42.50</td>
 <td>73.9 M</td>
 <td rowspan="1">
-SVTRv2 is a server-side text recognition model developed by the OpenOCR team from Fudan University's Vision and Learning Laboratory (FVL). It won the first prize in the PaddleOCR Algorithm Model Challenge - Task 1: OCR End-to-End Recognition Task, with a 6% improvement in end-to-end recognition accuracy compared to PP-OCRv4.
+SVTRv2 is a server text recognition model developed by the OpenOCR team of Fudan University's Visual and Learning Laboratory (FVL). It won the first prize in the PaddleOCR Algorithm Model Challenge - Task One: OCR End-to-End Recognition Task. The end-to-end recognition accuracy on the A list is 6% higher than that of PP-OCRv4.
 </td>
 </tr>
 </table>
@@ -176,20 +242,142 @@ SVTRv2 is a server-side text recognition model developed by the OpenOCR team fro
 <tr>
 <th>Model</th><th>Model Download Link</th>
 <th>Recognition Avg Accuracy(%)</th>
-<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>
 </tr>
 <tr>
-<td>ch_RepSVTR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0b2/ch_RepSVTR_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_RepSVTR_rec_pretrained.pdparams">Training Model</a></td>
+<td>ch_RepSVTR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_RepSVTR_rec_infer.tar">Inference Model</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_RepSVTR_rec_pretrained.pdparams">Training Model</a></td>
 <td>65.07</td>
-<td>5.93 / 5.93</td>
+<td>5.93 / 1.62</td>
 <td>20.73 / 7.32</td>
 <td>22.1 M</td>
-<td rowspan="1">RepSVTR is a mobile text recognition model based on SVTRv2. It won the first prize in the PaddleOCR Algorithm Model Challenge - Task 1: OCR End-to-End Recognition Task, with a 2.5% improvement in end-to-end recognition accuracy compared to PP-OCRv4 and comparable inference speed.</td>
+<td rowspan="1">    The RepSVTR text recognition model is a mobile text recognition model based on SVTRv2. It won the first prize in the PaddleOCR Algorithm Model Challenge - Task One: OCR End-to-End Recognition Task. The end-to-end recognition accuracy on the B list is 2.5% higher than that of PP-OCRv4, with the same inference speed.</td>
 </tr>
 </table>
+
+* <b>English Recognition Model</b>
+<table>
+<tr>
+<th>Model</th><th>Model Download Link</th>
+<th>Recognition Avg Accuracy(%)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>Model Storage Size (M)</th>
+<th>Introduction</th>
+</tr>
+<tr>
+<td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv4_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td> 70.39</td>
+<td>4.81 / 0.75</td>
+<td>16.10 / 5.31</td>
+<td>6.8 M</td>
+<td>The ultra-lightweight English recognition model trained based on the PP-OCRv4 recognition model supports the recognition of English and numbers.</td>
+</tr>
+<tr>
+<td>en_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/en_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>70.69</td>
+<td>5.44 / 0.75</td>
+<td>8.65 / 5.57</td>
+<td>7.8 M </td>
+<td>The ultra-lightweight English recognition model trained based on the PP-OCRv3 recognition model supports the recognition of English and numbers.</td>
+</tr>
+</table>
+
+* <b>Multilingual Recognition Model</b>
+<table>
+<tr>
+<th>Model</th><th>Model Download Link</th>
+<th>Recognition Avg Accuracy(%)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>CPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
+<th>Model Storage Size (M)</th>
+<th>Introduction</th>
+</tr>
+<tr>
+<td>korean_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/korean_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>60.21</td>
+<td>5.40 / 0.97</td>
+<td>9.11 / 4.05</td>
+<td>8.6 M</td>
+<td>The ultra-lightweight Korean recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Korean and numbers. </td>
+</tr>
+<tr>
+<td>japan_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/japan_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>45.69</td>
+<td>5.70 / 1.02</td>
+<td>8.48 / 4.07</td>
+<td>8.8 M </td>
+<td>The ultra-lightweight Japanese recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Japanese and numbers.</td>
+</tr>
+<tr>
+<td>chinese_cht_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/chinese_cht_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>82.06</td>
+<td>5.90 / 1.28</td>
+<td>9.28 / 4.34</td>
+<td>9.7 M </td>
+<td>The ultra-lightweight Traditional Chinese recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Traditional Chinese and numbers.</td>
+</tr>
+<tr>
+<td>te_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/te_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>95.88</td>
+<td>5.42 / 0.82</td>
+<td>8.10 / 6.91</td>
+<td>7.8 M </td>
+<td>The ultra-lightweight Telugu recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Telugu and numbers.</td>
+</tr>
+<tr>
+<td>ka_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ka_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>96.96</td>
+<td>5.25 / 0.79</td>
+<td>9.09 / 3.86</td>
+<td>8.0 M </td>
+<td>The ultra-lightweight Kannada recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Kannada and numbers.</td>
+</tr>
+<tr>
+<td>ta_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ta_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>76.83</td>
+<td>5.23 / 0.75</td>
+<td>10.13 / 4.30</td>
+<td>8.0 M </td>
+<td>The ultra-lightweight Tamil recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Tamil and numbers.</td>
+</tr>
+<tr>
+<td>latin_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/latin_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>76.93</td>
+<td>5.20 / 0.79</td>
+<td>8.83 / 7.15</td>
+<td>7.8 M</td>
+<td>The ultra-lightweight Latin recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Latin script and numbers.</td>
+</tr>
+<tr>
+<td>arabic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/arabic_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>73.55</td>
+<td>5.35 / 0.79</td>
+<td>8.80 / 4.56</td>
+<td>7.8 M</td>
+<td>The ultra-lightweight Arabic script recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Arabic script and numbers.</td>
+</tr>
+<tr>
+<td>cyrillic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/cyrillic_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>94.28</td>
+<td>5.23 / 0.76</td>
+<td>8.89 / 3.88</td>
+<td>7.9 M  </td>
+<td>
+The ultra-lightweight cyrillic alphabet recognition model trained based on the PP-OCRv3 recognition model supports the recognition of cyrillic letters and numbers.</td>
+</tr>
+<tr>
+<td>devanagari_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/devanagari_PP-OCRv3_mobile_rec_infer.tar">Inference Model</a>/<a href="">Training Model</a></td>
+<td>96.44</td>
+<td>5.22 / 0.79</td>
+<td>8.56 / 4.06</td>
+<td>7.9 M  </td>
+<td>The ultra-lightweight Devanagari script recognition model trained based on the PP-OCRv3 recognition model supports the recognition of Devanagari script and numbers.</td>
+</tr>
+</table>
+</details>
 <p><b>Layout Region Detection Module Models (Optional):</b></p>
 <table>
 <thead>
@@ -231,7 +419,7 @@ SVTRv2 is a server-side text recognition model developed by the OpenOCR team fro
 </table>
 
 > ❗ The above list includes the <b>3 core models</b> that are the focus of the layout detection module. The module supports a total of <b>11 full models</b>, including multiple predefined models with different categories. The complete list of models is as follows:
-
+<details><summary> 👉Model List Details</summary>
 * <b>Table Layout Detection Models</b>
 <table>
 <thead>
@@ -357,7 +545,7 @@ SVTRv2 is a server-side text recognition model developed by the OpenOCR team fro
 </tr>
 </tbody>
 </table>
-
+</details>
 <p><b>Text Image Correction Module Model (Optional):</b></p>
 <table>
 <thead>

docs/pipeline_usage/tutorials/ocr_pipelines/table_recognition_v2.md

@@ -131,19 +131,88 @@ comments: true
 <th>介绍</th>
 </tr>
 <tr>
+<td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_doc_pretrained.pdparams">训练模型</a></td>
+<td>81.53</td>
+<td>6.65 / 2.38</td>
+<td>32.92 / 32.92</td>
+<td>74.7 M</td>
+<td>PP-OCRv4_server_rec_doc是在PP-OCRv4_server_rec的基础上，在更多中文文档数据和PP-OCR训练数据的混合数据训练而成，增加了部分繁体字、日文、特殊字符的识别能力，可支持识别的字符为1.5万+，除文档相关的文字识别能力提升外，也同时提升了通用文字的识别能力</td>
+</tr>
+<tr>
 <td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
-<td>78.20</td>
-<td>4.82 / 4.82</td>
+<td>78.74</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
-<td rowspan="2">PP-OCRv4是百度飞桨视觉团队自研的文本识别模型PP-OCRv3的下一个版本，通过引入数据增强方案、GTC-NRTR指导分支等策略，在模型推理速度不变的情况下，进一步提升了文本识别精度。该模型提供了服务端（server）和移动端（mobile）两个不同版本，来满足不同场景下的工业需求。</td>
+<td>PP-OCRv4的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
 </tr>
 <tr>
 <td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">训练模型</a></td>
-<td>79.20</td>
-<td>6.58 / 6.58</td>
+<td>80.61 </td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
+<td>PP-OCRv4的服务器端模型，推理精度高，可以部署在多种不同的服务器上</td>
+</tr>
+<tr>
+<td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>70.39</td>
+<td>4.81 / 0.75</td>
+<td>16.10 / 5.31</td>
+<td>6.8 M</td>
+<td>基于PP-OCRv4识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
+</tr>
+</table>
+
+> ❗ 以上列出的是文本识别模块重点支持的<b>4个核心模型</b>，该模块总共支持<b>18个全量模型</b>，包含多个多语言文本识别模型，完整的模型列表如下：
+
+<details><summary> 👉模型列表详情</summary>
+
+* <b>中文识别模型</b>
+<table>
+<tr>
+<th>模型</th><th>模型下载链接</th>
+<th>识别 Avg Accuracy(%)</th>
+<th>GPU推理耗时（ms）<br/>[常规模式 / 高性能模式]</th>
+<th>CPU推理耗时（ms）<br/>[常规模式 / 高性能模式]</th>
+<th>模型存储大小（M）</th>
+<th>介绍</th>
+</tr>
+<tr>
+<td>PP-OCRv4_server_rec_doc</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_doc_pretrained.pdparams">训练模型</a></td>
+<td>81.53</td>
+<td>6.65 / 2.38</td>
+<td>32.92 / 32.92</td>
+<td>74.7 M</td>
+<td>PP-OCRv4_server_rec_doc是在PP-OCRv4_server_rec的基础上，在更多中文文档数据和PP-OCR训练数据的混合数据训练而成，增加了部分繁体字、日文、特殊字符的识别能力，可支持识别的字符为1.5万+，除文档相关的文字识别能力提升外，也同时提升了通用文字的识别能力</td>
+</tr>
+<tr>
+<td>PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>78.74</td>
+<td>4.82 / 1.20</td>
+<td>16.74 / 4.64</td>
+<td>10.6 M</td>
+<td>PP-OCRv4的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
+</tr>
+<tr>
+<td>PP-OCRv4_server_rec </td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/PP-OCRv4_server_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv4_server_rec_pretrained.pdparams">训练模型</a></td>
+<td>80.61 </td>
+<td>6.58 / 2.43</td>
+<td>33.17 / 33.17</td>
+<td>71.2 M</td>
+<td>PP-OCRv4的服务器端模型，推理精度高，可以部署在多种不同的服务器上</td>
+</tr>
+<tr>
+<td>PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>72.96</td>
+<td>5.87 / 1.19</td>
+<td>9.07 / 4.28</td>
+<td>9.2 M</td>
+<td>PP-OCRv3的轻量级识别模型，推理效率高，可以部署在包含端侧设备的多种硬件设备中</td>
 </tr>
 </table>
 
@@ -159,7 +228,7 @@ comments: true
 <tr>
 <td>ch_SVTRv2_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_SVTRv2_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_SVTRv2_rec_pretrained.pdparams">训练模型</a></td>
 <td>68.81</td>
-<td>8.08 / 8.08</td>
+<td>8.08 / 2.74</td>
 <td>50.17 / 42.50</td>
 <td>73.9 M</td>
 <td rowspan="1">
@@ -180,13 +249,146 @@ SVTRv2 是一种由复旦大学视觉与学习实验室（FVL）的OpenOCR团队
 <tr>
 <td>ch_RepSVTR_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/ch_RepSVTR_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ch_RepSVTR_rec_pretrained.pdparams">训练模型</a></td>
 <td>65.07</td>
-<td>5.93 / 5.93</td>
+<td>5.93 / 1.62</td>
 <td>20.73 / 7.32</td>
 <td>22.1 M</td>
 <td rowspan="1">    RepSVTR 文本识别模型是一种基于SVTRv2 的移动端文本识别模型，其在PaddleOCR算法模型挑战赛 - 赛题一：OCR端到端识别任务中荣获一等奖，B榜端到端识别精度相比PP-OCRv4提升2.5%，推理速度持平。</td>
 </tr>
 </table>
 
+* <b>英文识别模型</b>
+<table>
+<tr>
+<th>模型</th><th>模型下载链接</th>
+<th>识别 Avg Accuracy(%)</th>
+<th>GPU推理耗时（ms）<br/>[常规模式 / 高性能模式]</th>
+<th>CPU推理耗时（ms）<br/>[常规模式 / 高性能模式]</th>
+<th>模型存储大小（M）</th>
+<th>介绍</th>
+</tr>
+<tr>
+<td>en_PP-OCRv4_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv4_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td> 70.39</td>
+<td>4.81 / 0.75</td>
+<td>16.10 / 5.31</td>
+<td>6.8 M</td>
+<td>基于PP-OCRv4识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
+</tr>
+<tr>
+<td>en_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+en_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/en_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>70.69</td>
+<td>5.44 / 0.75</td>
+<td>8.65 / 5.57</td>
+<td>7.8 M </td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量英文识别模型，支持英文、数字识别</td>
+</tr>
+</table>
+
+
+* <b>多语言识别模型</b>
+<table>
+<tr>
+<th>模型</th><th>模型下载链接</th>
+<th>识别 Avg Accuracy(%)</th>
+<th>GPU推理耗时（ms）<br/>[常规模式 / 高性能模式]</th>
+<th>CPU推理耗时（ms）<br/>[常规模式 / 高性能模式]</th>
+<th>模型存储大小（M）</th>
+<th>介绍</th>
+</tr>
+<tr>
+<td>korean_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+korean_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/korean_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>60.21</td>
+<td>5.40 / 0.97</td>
+<td>9.11 / 4.05</td>
+<td>8.6 M</td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量韩文识别模型，支持韩文、数字识别</td>
+</tr>
+<tr>
+<td>japan_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+japan_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/japan_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>45.69</td>
+<td>5.70 / 1.02</td>
+<td>8.48 / 4.07</td>
+<td>8.8 M </td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量日文识别模型，支持日文、数字识别</td>
+</tr>
+<tr>
+<td>chinese_cht_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+chinese_cht_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/chinese_cht_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>82.06</td>
+<td>5.90 / 1.28</td>
+<td>9.28 / 4.34</td>
+<td>9.7 M </td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量繁体中文识别模型，支持繁体中文、数字识别</td>
+</tr>
+<tr>
+<td>te_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+te_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/te_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>95.88</td>
+<td>5.42 / 0.82</td>
+<td>8.10 / 6.91</td>
+<td>7.8 M </td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量泰卢固文识别模型，支持泰卢固文、数字识别</td>
+</tr>
+<tr>
+<td>ka_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+ka_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ka_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>96.96</td>
+<td>5.25 / 0.79</td>
+<td>9.09 / 3.86</td>
+<td>8.0 M </td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量卡纳达文识别模型，支持卡纳达文、数字识别</td>
+</tr>
+<tr>
+<td>ta_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+ta_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/ta_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>76.83</td>
+<td>5.23 / 0.75</td>
+<td>10.13 / 4.30</td>
+<td>8.0 M </td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量泰米尔文识别模型，支持泰米尔文、数字识别</td>
+</tr>
+<tr>
+<td>latin_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+latin_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/latin_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>76.93</td>
+<td>5.20 / 0.79</td>
+<td>8.83 / 7.15</td>
+<td>7.8 M</td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量拉丁文识别模型，支持拉丁文、数字识别</td>
+</tr>
+<tr>
+<td>arabic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+arabic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/arabic_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>73.55</td>
+<td>5.35 / 0.79</td>
+<td>8.80 / 4.56</td>
+<td>7.8 M</td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量阿拉伯字母识别模型，支持阿拉伯字母、数字识别</td>
+</tr>
+<tr>
+<td>cyrillic_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+cyrillic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/cyrillic_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>94.28</td>
+<td>5.23 / 0.76</td>
+<td>8.89 / 3.88</td>
+<td>7.9 M  </td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量斯拉夫字母识别模型，支持斯拉夫字母、数字识别</td>
+</tr>
+<tr>
+<td>devanagari_PP-OCRv3_mobile_rec</td><td><a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_inference_model/paddle3.0rc0/\
+devanagari_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/devanagari_PP-OCRv3_mobile_rec_pretrained.pdparams">训练模型</a></td>
+<td>96.44</td>
+<td>5.22 / 0.79</td>
+<td>8.56 / 4.06</td>
+<td>7.9 M</td>
+<td>基于PP-OCRv3识别模型训练得到的超轻量梵文字母识别模型，支持梵文字母、数字识别</td>
+</tr>
+</table>
+</details>
 <table>
 <thead>
 <tr>
@@ -227,7 +429,7 @@ SVTRv2 是一种由复旦大学视觉与学习实验室（FVL）的OpenOCR团队
 </table>
 
 > ❗ 以上列出的是版面检测模块重点支持的<b>3个核心模型</b>，该模块总共支持<b>11个全量模型</b>，包含多个预定义了不同类别的模型，完整的模型列表如下：
-
+<details><summary> 👉模型列表详情</summary>
 * <b>表格版面检测模型</b>
 <table>
 <thead>
@@ -372,7 +574,7 @@ SVTRv2 是一种由复旦大学视觉与学习实验室（FVL）的OpenOCR团队
 </tr>
 </tbody>
 </table>
-
+</details>
 <p><b>文档图像方向分类模块模型（可选）：</b></p>
 <table>
 <thead>

docs/practical_tutorials/document_scene_information_extraction(seal_recognition)_tutorial.en.md

@@ -88,7 +88,7 @@ PaddleX provides 2 end-to-end seal text detection models, which can be reference
 <tr>
 <th>Model</th>
 <th>mAP(0.5) (%)</th>
-<th>GPU Inference Time (ms)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU Inference Time (ms)</th>
 <th>Model Size (M)</th>
 <th>Description</th>

docs/practical_tutorials/image_classification_garbage_tutorial.en.md

@@ -42,7 +42,7 @@ PaddleX provides 80 end-to-end image classification models, which can be referen
 <tr>
 <th>Model List</th>
 <th>Top-1 Accuracy (%)</th>
-<th>GPU Inference Time (ms)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU Inference Time (ms)</th>
 <th>Model Size (M)</th>
 </tr>

docs/practical_tutorials/ocr_det_license_tutorial.en.md

@@ -40,7 +40,7 @@ PaddleX provides 4 end-to-end text detection models. For details, refer to the [
 <tr>
 <th>Model</th>
 <th>Detection Hmean (%)</th>
-<th>GPU Inference Time (ms)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU Inference Time (ms)</th>
 <th>Model Storage Size (M)</th>
 <th>Introduction</th>

docs/practical_tutorials/semantic_segmentation_road_tutorial.en.md

@@ -40,7 +40,7 @@ PaddleX provides 18 end-to-end semantic segmentation models. For details, refer
 <tr>
 <th>Model List</th>
 <th>mIoU (%)</th>
-<th>GPU Inference Time (ms)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU Inference Time (ms)</th>
 <th>Model Size (M)</th>
 </tr>

docs/practical_tutorials/small_object_detection_tutorial.en.md

@@ -45,7 +45,7 @@ PaddleX provides 3 high-precision and high-efficiency end-to-end small object de
 <th>Model</th><th>Model Download Link</th>
 <th>mAP(0.5:0.95)</th>
 <th>mAP(0.5)</th>
-<th>GPU Inference Time (ms)</th>
+<th>GPU Inference Time (ms)<br/>[Normal Mode / High-Performance Mode]</th>
 <th>CPU Inference Time (ms)</th>
 <th>Model Storage Size (M)</th>
 <th>Description</th>

docs/support_list/models_list.en.md

@@ -1568,8 +1568,6 @@ PaddleX includes multiple production lines, each containing several modules, and
 
 ## [Human Keypoint Detection Module](../module_usage/tutorials//cv_modules/human_keypoint_detection.en.md)
 
-## [Human Keypoint Detection Module](../module_usage/tutorials//cv_modules/human_keypoint_detection.md)
-
 <table>
 <tr>
 <th>Model</th>
@@ -2071,7 +2069,7 @@ PaddleX includes multiple production lines, each containing several modules, and
 <tr>
 <td>PP-OCRv4_server_rec_doc</td>
 <td>81.53</td>
-<td>6.65 / 6.65</td>
+<td>6.65 / 2.38</td>
 <td>32.92 / 32.92</td>
 <td>74.7 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/PP-OCRv4_server_rec_doc.yaml">PP-OCRv4_server_rec_doc.yaml</a></td>
@@ -2080,7 +2078,7 @@ PaddleX includes multiple production lines, each containing several modules, and
 <tr>
 <td>PP-OCRv4_mobile_rec</td>
 <td>78.74</td>
-<td>4.82 / 4.82</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/PP-OCRv4_mobile_rec.yaml">PP-OCRv4_mobile_rec.yaml</a></td>
@@ -2089,7 +2087,7 @@ PaddleX includes multiple production lines, each containing several modules, and
 <tr>
 <td>PP-OCRv4_server_rec </td>
 <td>80.61 </td>
-<td>6.58 / 6.58</td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/PP-OCRv4_server_rec.yaml">PP-OCRv4_server_rec.yaml</a></td>
@@ -2098,7 +2096,7 @@ PaddleX includes multiple production lines, each containing several modules, and
 <tr>
 <td>PP-OCRv3_mobile_rec</td>
 <td>72.96</td>
-<td>5.87 / 5.87</td>
+<td>5.87 / 1.19</td>
 <td>9.07 / 4.28</td>
 <td>9.2 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/PP-OCRv3_mobile_rec.yaml">PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2119,7 +2117,7 @@ PaddleX includes multiple production lines, each containing several modules, and
 <tr>
 <td>ch_SVTRv2_rec</td>
 <td>68.81</td>
-<td>8.08 / 8.08</td>
+<td>8.08 / 2.74</td>
 <td>50.17 / 42.50</td>
 <td>73.9 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/ch_SVTRv2_rec.yaml">ch_SVTRv2_rec.yaml</a></td>
@@ -2140,7 +2138,7 @@ PaddleX includes multiple production lines, each containing several modules, and
 <tr>
 <td>ch_RepSVTR_rec</td>
 <td>65.07</td>
-<td>5.93 / 5.93</td>
+<td>5.93 / 1.62</td>
 <td>20.73 / 7.32</td>
 <td>22.1 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/ch_RepSVTR_rec.yaml">ch_RepSVTR_rec.yaml</a></td>
@@ -2162,7 +2160,7 @@ PaddleX includes multiple production lines, each containing several modules, and
 <tr>
 <td>en_PP-OCRv4_mobile_rec</td>
 <td> 70.39</td>
-<td>4.81 / 4.81</td>
+<td>4.81 / 0.75</td>
 <td>16.10 / 5.31</td>
 <td>6.8 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/en_PP-OCRv4_mobile_rec.yaml">en_PP-OCRv4_mobile_rec.yaml</a></td>
@@ -2171,7 +2169,7 @@ PaddleX includes multiple production lines, each containing several modules, and
 <tr>
 <td>en_PP-OCRv3_mobile_rec</td>
 <td>70.69</td>
-<td>5.44 / 5.44</td>
+<td>5.44 / 0.75</td>
 <td>8.65 / 5.57</td>
 <td>7.8 M </td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/en_PP-OCRv3_mobile_rec.yaml">en_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2195,7 +2193,7 @@ PaddleX includes multiple production lines, each containing several modules, and
 <tr>
 <td>korean_PP-OCRv3_mobile_rec</td>
 <td>60.21</td>
-<td>5.40 / 5.40</td>
+<td>5.40 / 0.97</td>
 <td>9.11 / 4.05</td>
 <td>8.6 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/korean_PP-OCRv3_mobile_rec.yaml">korean_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2204,7 +2202,7 @@ PaddleX includes multiple production lines, each containing several modules, and
 <tr>
 <td>japan_PP-OCRv3_mobile_rec</td>
 <td>45.69</td>
-<td>5.70 / 5.70</td>
+<td>5.70 / 1.02</td>
 <td>8.48 / 4.07</td>
 <td>8.8 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/japan_PP-OCRv3_mobile_rec.yaml">japan_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2213,7 +2211,7 @@ PaddleX includes multiple production lines, each containing several modules, and
 <tr>
 <td>chinese_cht_PP-OCRv3_mobile_rec</td>
 <td>82.06</td>
-<td>5.90 / 5.90</td>
+<td>5.90 / 1.28</td>
 <td>9.28 / 4.34</td>
 <td>9.7 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/chinese_cht_PP-OCRv3_mobile_rec.yaml">chinese_cht_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2222,7 +2220,7 @@ PaddleX includes multiple production lines, each containing several modules, and
 <tr>
 <td>te_PP-OCRv3_mobile_rec</td>
 <td>95.88</td>
-<td>5.42 / 5.42</td>
+<td>5.42 / 0.82</td>
 <td>8.10 / 6.91</td>
 <td>7.8 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/te_PP-OCRv3_mobile_rec.yaml">te_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2231,7 +2229,7 @@ PaddleX includes multiple production lines, each containing several modules, and
 <tr>
 <td>ka_PP-OCRv3_mobile_rec</td>
 <td>96.96</td>
-<td>5.25 / 5.25</td>
+<td>5.25 / 0.79</td>
 <td>9.09 / 3.86</td>
 <td>8.0 M </td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/ka_PP-OCRv3_mobile_rec.yaml">ka_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2240,7 +2238,7 @@ PaddleX includes multiple production lines, each containing several modules, and
 <tr>
 <td>ta_PP-OCRv3_mobile_rec</td>
 <td>76.83</td>
-<td>5.23 / 5.23</td>
+<td>5.23 / 0.75</td>
 <td>10.13 / 4.30</td>
 <td>8.0 M </td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/ta_PP-OCRv3_mobile_rec.yaml">ta_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2249,7 +2247,7 @@ PaddleX includes multiple production lines, each containing several modules, and
 <tr>
 <td>latin_PP-OCRv3_mobile_rec</td>
 <td>76.93</td>
-<td>5.20 / 5.20</td>
+<td>5.20 / 0.79</td>
 <td>8.83 / 7.15</td>
 <td>7.8 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/latin_PP-OCRv3_mobile_rec.yaml">latin_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2258,7 +2256,7 @@ PaddleX includes multiple production lines, each containing several modules, and
 <tr>
 <td>arabic_PP-OCRv3_mobile_rec</td>
 <td>73.55</td>
-<td>5.35 / 5.35</td>
+<td>5.35 / 0.79</td>
 <td>8.80 / 4.56</td>
 <td>7.8 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/arabic_PP-OCRv3_mobile_rec.yaml">arabic_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2267,7 +2265,7 @@ PaddleX includes multiple production lines, each containing several modules, and
 <tr>
 <td>cyrillic_PP-OCRv3_mobile_rec</td>
 <td>94.28</td>
-<td>5.23 / 5.23</td>
+<td>5.23 / 0.76</td>
 <td>8.89 / 3.88</td>
 <td>7.9 M  </td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/cyrillic_PP-OCRv3_mobile_rec.yaml">cyrillic_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2276,7 +2274,7 @@ PaddleX includes multiple production lines, each containing several modules, and
 <tr>
 <td>devanagari_PP-OCRv3_mobile_rec</td>
 <td>96.44</td>
-<td>5.22 / 5.22</td>
+<td>5.22 / 0.79</td>
 <td>8.56 / 4.06</td>
 <td>7.9 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/devanagari_PP-OCRv3_mobile_rec.yaml">devanagari_PP-OCRv3_mobile_rec.yaml</a></td>

docs/support_list/models_list.md

@@ -1929,7 +1929,7 @@ PaddleX 内置了多条产线，每条产线都包含了若干模块，每个模
 <tr>
 <td>PP-OCRv4_server_rec_doc</td>
 <td>81.53</td>
-<td>6.65 / 6.65</td>
+<td>6.65 / 2.38</td>
 <td>32.92 / 32.92</td>
 <td>74.7 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/PP-OCRv4_server_rec_doc.yaml">PP-OCRv4_server_rec_doc.yaml</a></td>
@@ -1939,7 +1939,7 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a><
 <tr>
 <td>PP-OCRv4_mobile_rec</td>
 <td>78.74</td>
-<td>4.82 / 4.82</td>
+<td>4.82 / 1.20</td>
 <td>16.74 / 4.64</td>
 <td>10.6 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/PP-OCRv4_mobile_rec.yaml">PP-OCRv4_mobile_rec.yaml</a></td>
@@ -1948,7 +1948,7 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a><
 <tr>
 <td>PP-OCRv4_server_rec </td>
 <td>80.61 </td>
-<td>6.58 / 6.58</td>
+<td>6.58 / 2.43</td>
 <td>33.17 / 33.17</td>
 <td>71.2 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/PP-OCRv4_server_rec.yaml">PP-OCRv4_server_rec.yaml</a></td>
@@ -1957,7 +1957,7 @@ PP-OCRv4_server_rec_doc_infer.tar">推理模型</a>/<a href="">训练模型</a><
 <tr>
 <td>PP-OCRv3_mobile_rec</td>
 <td>72.96</td>
-<td>5.87 / 5.87</td>
+<td>5.87 / 1.19</td>
 <td>9.07 / 4.28</td>
 <td>9.2 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/PP-OCRv3_mobile_rec.yaml">PP-OCRv3_mobile_rec.yaml</a></td>
@@ -1979,7 +1979,7 @@ PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
 <tr>
 <td>ch_SVTRv2_rec</td>
 <td>68.81</td>
-<td>8.08 / 8.08</td>
+<td>8.08 / 2.74</td>
 <td>50.17 / 42.50</td>
 <td>73.9 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/ch_SVTRv2_rec.yaml">ch_SVTRv2_rec.yaml</a></td>
@@ -2000,7 +2000,7 @@ PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
 <tr>
 <td>ch_RepSVTR_rec</td>
 <td>65.07</td>
-<td>5.93 / 5.93</td>
+<td>5.93 / 1.62</td>
 <td>20.73 / 7.32</td>
 <td>22.1 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/ch_RepSVTR_rec.yaml">ch_RepSVTR_rec.yaml</a></td>
@@ -2023,7 +2023,7 @@ PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></td>
 <tr>
 <td>en_PP-OCRv4_mobile_rec</td>
 <td> 70.39</td>
-<td>4.81 / 4.81</td>
+<td>4.81 / 0.75</td>
 <td>16.10 / 5.31</td>
 <td>6.8 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/en_PP-OCRv4_mobile_rec.yaml">en_PP-OCRv4_mobile_rec.yaml</a></td>
@@ -2033,7 +2033,7 @@ en_PP-OCRv4_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></
 <tr>
 <td>en_PP-OCRv3_mobile_rec</td>
 <td>70.69</td>
-<td>5.44 / 5.44</td>
+<td>5.44 / 0.75</td>
 <td>8.65 / 5.57</td>
 <td>7.8 M </td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/en_PP-OCRv3_mobile_rec.yaml">en_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2058,7 +2058,7 @@ en_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></
 <tr>
 <td>korean_PP-OCRv3_mobile_rec</td>
 <td>60.21</td>
-<td>5.40 / 5.40</td>
+<td>5.40 / 0.97</td>
 <td>9.11 / 4.05</td>
 <td>8.6 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/korean_PP-OCRv3_mobile_rec.yaml">korean_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2068,7 +2068,7 @@ korean_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</
 <tr>
 <td>japan_PP-OCRv3_mobile_rec</td>
 <td>45.69</td>
-<td>5.70 / 5.70</td>
+<td>5.70 / 1.02</td>
 <td>8.48 / 4.07</td>
 <td>8.8 M </td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/japan_PP-OCRv3_mobile_rec.yaml">japan_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2078,7 +2078,7 @@ japan_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a
 <tr>
 <td>chinese_cht_PP-OCRv3_mobile_rec</td>
 <td>82.06</td>
-<td>5.90 / 5.90</td>
+<td>5.90 / 1.28</td>
 <td>9.28 / 4.34</td>
 <td>9.7 M </td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/chinese_cht_PP-OCRv3_mobile_rec.yaml">chinese_cht_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2088,7 +2088,7 @@ chinese_cht_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模
 <tr>
 <td>te_PP-OCRv3_mobile_rec</td>
 <td>95.88</td>
-<td>5.42 / 5.42</td>
+<td>5.42 / 0.82</td>
 <td>8.10 / 6.91</td>
 <td>7.8 M </td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/te_PP-OCRv3_mobile_rec.yaml">te_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2098,7 +2098,7 @@ te_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></
 <tr>
 <td>ka_PP-OCRv3_mobile_rec</td>
 <td>96.96</td>
-<td>5.25 / 5.25</td>
+<td>5.25 / 0.79</td>
 <td>9.09 / 3.86</td>
 <td>8.0 M </td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/ka_PP-OCRv3_mobile_rec.yaml">ka_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2108,7 +2108,7 @@ ka_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></
 <tr>
 <td>ta_PP-OCRv3_mobile_rec</td>
 <td>76.83</td>
-<td>5.23 / 5.23</td>
+<td>5.23 / 0.75</td>
 <td>10.13 / 4.30</td>
 <td>8.0 M </td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/ta_PP-OCRv3_mobile_rec.yaml">ta_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2118,7 +2118,7 @@ ta_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a></
 <tr>
 <td>latin_PP-OCRv3_mobile_rec</td>
 <td>76.93</td>
-<td>5.20 / 5.20</td>
+<td>5.20 / 0.79</td>
 <td>8.83 / 7.15</td>
 <td>7.8 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/latin_PP-OCRv3_mobile_rec.yaml">latin_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2128,7 +2128,7 @@ latin_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</a
 <tr>
 <td>arabic_PP-OCRv3_mobile_rec</td>
 <td>73.55</td>
-<td>5.35 / 5.35</td>
+<td>5.35 / 0.79</td>
 <td>8.80 / 4.56</td>
 <td>7.8 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/arabic_PP-OCRv3_mobile_rec.yaml">arabic_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2138,7 +2138,7 @@ arabic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型</
 <tr>
 <td>cyrillic_PP-OCRv3_mobile_rec</td>
 <td>94.28</td>
-<td>5.23 / 5.23</td>
+<td>5.23 / 0.76</td>
 <td>8.89 / 3.88</td>
 <td>7.9 M  </td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/cyrillic_PP-OCRv3_mobile_rec.yaml">cyrillic_PP-OCRv3_mobile_rec.yaml</a></td>
@@ -2148,7 +2148,7 @@ cyrillic_PP-OCRv3_mobile_rec_infer.tar">推理模型</a>/<a href="">训练模型
 <tr>
 <td>devanagari_PP-OCRv3_mobile_rec</td>
 <td>96.44</td>
-<td>5.22 / 5.22</td>
+<td>5.22 / 0.79</td>
 <td>8.56 / 4.06</td>
 <td>7.9 M</td>
 <td><a href="https://github.com/PaddlePaddle/PaddleX/blob/develop/paddlex/configs/modules/text_recognition/devanagari_PP-OCRv3_mobile_rec.yaml">devanagari_PP-OCRv3_mobile_rec.yaml</a></td>