diff --git a/src/transformers/models/siglip/modeling_siglip.py b/src/transformers/models/siglip/modeling_siglip.py
index 252315d4186..d6966c31f66 100644
--- a/src/transformers/models/siglip/modeling_siglip.py
+++ b/src/transformers/models/siglip/modeling_siglip.py
@@ -370,7 +370,7 @@ def eager_attention_forward(
 class SiglipAttention(nn.Module):
     """Multi-headed attention from 'Attention Is All You Need' paper"""
 
-    def __init__(self, config: Union[SiglipVisionConfig, SiglipTextConfig]):
+    def __init__(self, config):
         super().__init__()
         self.config = config
         self.embed_dim = config.hidden_size
diff --git a/src/transformers/models/siglip2/modeling_siglip2.py b/src/transformers/models/siglip2/modeling_siglip2.py
index 8614384241b..a198cbc347f 100644
--- a/src/transformers/models/siglip2/modeling_siglip2.py
+++ b/src/transformers/models/siglip2/modeling_siglip2.py
@@ -264,7 +264,7 @@ def eager_attention_forward(
 class Siglip2Attention(nn.Module):
     """Multi-headed attention from 'Attention Is All You Need' paper"""
 
-    def __init__(self, config: Union[Siglip2VisionConfig, Siglip2TextConfig]):
+    def __init__(self, config):
         super().__init__()
         self.config = config
         self.embed_dim = config.hidden_size
diff --git a/tests/test_modeling_common.py b/tests/test_modeling_common.py
index 9ee6a93dbbc..9f56c79956e 100755
--- a/tests/test_modeling_common.py
+++ b/tests/test_modeling_common.py
@@ -739,7 +739,6 @@ class ModelTesterMixin:
             model = model_class(config)
             model.to(torch_device)
             model.eval()
-            print(model_class)
             with torch.no_grad():
                 first = model(**self._prepare_for_class(inputs_dict, model_class))[0]
                 second = model(**self._prepare_for_class(inputs_dict, model_class))[0]