diff --git a/README.md b/README.md
index bf3ee98..d4c8910 100644
--- a/README.md
+++ b/README.md
@@ -6,6 +6,8 @@ We introduce a audio2face model for realtime 2D chat avatar, which can run in 30
 - A lightweight 2D face generator model for mouth movement rendering, which can also be deployed on mobile devices realizing realtime inference.
 ## Data Preparation
 Get sample avatar data located in `./data/sample_data.zip` and extract to you path
+
+🔥More avatars can be found at [LiteAvatarGallery](https://modelscope.cn/models/HumanAIGC-Engineering/LiteAvatarGallery/summary)
 ## Installation
 We recommend a python version = 3.10 and cuda version = 11.8. Then build environment as follows:
 ```shell
@@ -42,4 +44,4 @@ If you find this project useful, please ⭐️ star the repository and cite our
   pages        = {5956--5959},
   year         = {2022}
 }
-```
\ No newline at end of file
+```
diff --git a/lite_avatar.py b/lite_avatar.py
index 2907412..ead894e 100644
--- a/lite_avatar.py
+++ b/lite_avatar.py
@@ -365,4 +365,4 @@ if __name__ == '__main__':
     lite_avatar = liteAvatar(data_dir=args.data_dir, num_threads=1, generate_offline=True)
     
     lite_avatar.handle(audio_file, tmp_frame_dir)
-    
\ No newline at end of file
+