diff --git a/modelcenter/PP-OCRv3/benchmark_cn.md b/modelcenter/PP-OCRv3/benchmark_cn.md
index b37769e99980e549f06d44ae2df743cf5665a198..b82157001b0e60c35fe96c49451d6cf3e58a357a 100644
--- a/modelcenter/PP-OCRv3/benchmark_cn.md
+++ b/modelcenter/PP-OCRv3/benchmark_cn.md
@@ -19,4 +19,4 @@
 
 
 ## 2. 相关使用说明
-1. https://github.com/PaddlePaddle/PaddleOCR/edit/dygraph/doc/doc_ch/PP-OCRv3_introduction.md
+1. https://github.com/PaddlePaddle/PaddleOCR/blob/dygraph/doc/doc_ch/PP-OCRv3_introduction.md
diff --git a/modelcenter/PP-OCRv3/benchmark_en.md b/modelcenter/PP-OCRv3/benchmark_en.md
index ce9f831879b7a8f7390661f571fca13480031b1c..4686907a5ab620f035c02bb8c738396d8598a412 100644
--- a/modelcenter/PP-OCRv3/benchmark_en.md
+++ b/modelcenter/PP-OCRv3/benchmark_en.md
@@ -15,4 +15,4 @@ The PP-OCRv3 model inference speed test uses 6148CPU, MKLDNN is turned on, and t
 | PP-OCRv3 | 62.90% | 15.6 | 331.00 |
 
 ## 2. Reference
-1. https://github.com/PaddlePaddle/PaddleOCR/edit/dygraph/doc/doc_ch/PP-OCRv3_introduction.md
+1. https://github.com/PaddlePaddle/PaddleOCR/blob/dygraph/doc/doc_en/PP-OCRv3_introduction_en.md