diff --git a/README.md b/README.md
index ec66ccba4326f9e2b39f56fe86c57e04eae91c19..2ffb92c94340d548f59468af9bc38bf0e0398b1b 100644
--- a/README.md
+++ b/README.md
@@ -7,7 +7,7 @@
 PaddleClas is a toolset for image classification tasks prepared for the industry and academia. It helps users train better computer vision models and apply them in real scenarios.
 
 **Recent update**
-- 2021.04.15 Add `MixNet` and `ReXNet` pretrained models, `MixNet`'s Top-1 Acc on ImageNet-1k reaches 78.6% and `ReXNet` reaches 82.09%.
+- 2021.04.15 Add `MixNet` and `ReXNet` pretrained models, `MixNet_L`'s Top-1 Acc on ImageNet-1k reaches 78.6% and `ReXNet_3_0` reaches 82.09%.
 - 2021.03.02 Add support for model quantization.
 - 2021.02.01 Add `RepVGG` pretrained models, whose Top-1 Acc on ImageNet-1k dataset reaches 79.65%.
 - 2021.01.27 Add `ViT` and `DeiT` pretrained models, `ViT`'s Top-1 Acc on ImageNet-1k dataset reaches 85.13%, and `DeiT` reaches 85.1%.
diff --git a/README_cn.md b/README_cn.md
index a0f8770031b1d099d60265a22575e51ea760258d..a0513dc79ca44f7f83c4294626e595d4bd65829a 100644
--- a/README_cn.md
+++ b/README_cn.md
@@ -8,7 +8,7 @@
 
 **近期更新**
 
-- 2021.04.15 添加`MixNet`和`ReXNet`系列模型，在ImageNet-1k上`MixNet` 模型Top1 Acc可达78.6%，`ReXNet`模型可达82.09%
+- 2021.04.15 添加`MixNet`和`ReXNet`系列模型，在ImageNet-1k上`MixNet_L` 模型Top1 Acc可达78.6%，`ReXNet_3_0`模型可达82.09%
 - 2021.03.02 添加分类模型量化方法与使用教程。
 - 2021.02.01 添加`RepVGG`系列模型，在ImageNet-1k上Top-1 Acc可达79.65%。
 - 2021.01.27 添加`ViT`与`DeiT`模型，在ImageNet-1k上，`ViT`模型Top-1 Acc可达85.13%，`DeiT`模型可达85.1%。
diff --git a/docs/en/update_history_en.md b/docs/en/update_history_en.md
index f045bcffd9d6422b28b3a19bac1d040a89b4561d..cef5384e424f944ce146bcccdbb25294885949cd 100644
--- a/docs/en/update_history_en.md
+++ b/docs/en/update_history_en.md
@@ -3,7 +3,7 @@
 
 
 - 2021.04.15
-   - Add `MixNet` and `ReXNet` pretrained models, `MixNet`'s Top-1 Acc on ImageNet-1k reaches 78.6% and `ReXNet` reaches 82.09%.
+   - Add `MixNet` and `ReXNet` pretrained models, `MixNet_L`'s Top-1 Acc on ImageNet-1k reaches 78.6% and `ReXNet_3_0` reaches 82.09%.
 - 2021.01.27
    * Add ViT and DeiT pretrained models, ViT's Top-1 Acc on ImageNet reaches 81.05%, and DeiT reaches 85.5%.
 - 2021.01.08
diff --git a/docs/zh_CN/update_history.md b/docs/zh_CN/update_history.md
index 23d853e2e0cb836eeaaa80232b77dfb8cdb24368..89f7b146dad555df0041994f41bcd1d9fed8131c 100644
--- a/docs/zh_CN/update_history.md
+++ b/docs/zh_CN/update_history.md
@@ -3,7 +3,7 @@
 
 
 - 2021.04.15
-   - 添加`MixNet`和`ReXNet`系列模型，在ImageNet-1k上`MixNet` 模型Top1 Acc可达78.6%，`ReXNet`模型可达82.09%
+   - 添加`MixNet_L`和`ReXNet_3_0`系列模型，在ImageNet-1k上`MixNet` 模型Top1 Acc可达78.6%，`ReXNet`模型可达82.09%
 - 2021.01.27
    * 添加ViT与DeiT模型，在ImageNet上，ViT模型Top-1 Acc可达81.05%，DeiT模型可达85.5%。
 - 2021.01.08
diff --git a/ppcls/modeling/architectures/mixnet.py b/ppcls/modeling/architectures/mixnet.py
index 0cdcecc56f02f76bde99272df728943bc2dcdd7a..f195694741292977e07b084a3c8efa8303f810a1 100644
--- a/ppcls/modeling/architectures/mixnet.py
+++ b/ppcls/modeling/architectures/mixnet.py
@@ -438,8 +438,7 @@ class MixUnit(nn.Layer):
     in_channels : int
         Number of input channels.
     out_channels : int
-        Number of output channels.
-    exp_channels : int
+        Number of output channels.  exp_channels : int
         Number of middle (expanded) channels.
     stride : int or tuple/list of 2 int
         Strides of the second convolution layer.
@@ -692,12 +691,7 @@ class MixNet(nn.Layer):
         return x
 
 
-def get_mixnet(version,
-               width_scale,
-               model_name=None,
-               pretrained=False,
-               root=os.path.join("~", ".paddle", "models"),
-               **kwargs):
+def get_mixnet(version, width_scale, model_name=None, **kwargs):
     """
     Create MixNet model with specific parameters.
 
@@ -708,11 +702,7 @@ def get_mixnet(version,
     width_scale : float
         Scale factor for width of layers.
     model_name : str or None, default None
-        Model name for loading pretrained model.
-    pretrained : bool, default False
-        Whether to load the pretrained weights for model.
-    root : str, default '~/.torch/models'
-        Location for keeping the model parameters.
+        Model name.
     """
 
     if version == "s":
@@ -769,13 +759,6 @@ def MixNet_S(**kwargs):
     """
     MixNet-S model from 'MixConv: Mixed Depthwise Convolutional Kernels,'
     https://arxiv.org/abs/1907.09595.
-
-    Parameters:
-    ----------
-    pretrained : bool, default False
-        Whether to load the pretrained weights for model.
-    root : str, default '~/.torch/models'
-        Location for keeping the model parameters.
     """
     return get_mixnet(
         version="s", width_scale=1.0, model_name="MixNet_S", **kwargs)
@@ -785,13 +768,6 @@ def MixNet_M(**kwargs):
     """
     MixNet-M model from 'MixConv: Mixed Depthwise Convolutional Kernels,'
     https://arxiv.org/abs/1907.09595.
-
-    Parameters:
-    ----------
-    pretrained : bool, default False
-        Whether to load the pretrained weights for model.
-    root : str, default '~/.torch/models'
-        Location for keeping the model parameters.
     """
     return get_mixnet(
         version="m", width_scale=1.0, model_name="MixNet_M", **kwargs)
@@ -801,13 +777,6 @@ def MixNet_L(**kwargs):
     """
     MixNet-L model from 'MixConv: Mixed Depthwise Convolutional Kernels,'
     https://arxiv.org/abs/1907.09595.
-
-    Parameters:
-    ----------
-    pretrained : bool, default False
-        Whether to load the pretrained weights for model.
-    root : str, default '~/.torch/models'
-        Location for keeping the model parameters.
     """
     return get_mixnet(
         version="m", width_scale=1.3, model_name="MixNet_L", **kwargs)