diff --git a/unsloth/models/llama.py b/unsloth/models/llama.py
index 888015b10e..3504037b66 100644
--- a/unsloth/models/llama.py
+++ b/unsloth/models/llama.py
@@ -1981,8 +1981,8 @@ class FastLlamaModel:
         # Patch generate
         if model.generate.__name__ != "unsloth_fast_generate":
             model._old_generate = model.generate
+            unsloth_fast_generate.__doc__ = model._old_generate.__doc__
             model.generate = types.MethodType(unsloth_fast_generate, model)
-            model.generate.__doc__ = model._old_generate.__doc__
         pass
         return model, tokenizer
     pass
@@ -2420,8 +2420,8 @@ class FastLlamaModel:
         # Patch generate
         if model.generate.__name__ != "unsloth_fast_generate":
             model._old_generate = model.generate
+            unsloth_fast_generate.__doc__ = model._old_generate.__doc__
             model.generate = types.MethodType(unsloth_fast_generate, model)
-            model.generate.__doc__ = model._old_generate.__doc__
         return model
     pass
 
diff --git a/unsloth/models/vision.py b/unsloth/models/vision.py
index 9eb7f6e99f..f249475edc 100644
--- a/unsloth/models/vision.py
+++ b/unsloth/models/vision.py
@@ -219,8 +219,8 @@ class FastBaseModel:
         # Patch generate
         if model.generate.__name__ != "unsloth_base_fast_generate":
             model._old_generate = model.generate
+            unsloth_base_fast_generate.__doc__ = model._old_generate.__doc__
             model.generate = types.MethodType(unsloth_base_fast_generate, model)
-            model.generate.__doc__ = model._old_generate.__doc__
         return model, tokenizer
     pass
 
@@ -371,8 +371,8 @@ class FastBaseModel:
         # Patch generate
         if model.generate.__name__ != "unsloth_base_fast_generate":
             model._old_generate = model.generate
+            unsloth_base_fast_generate.__doc__ = model._old_generate.__doc__
             model.generate = types.MethodType(unsloth_base_fast_generate, model)
-            model.generate.__doc__ = model._old_generate.__doc__
         return model
     pass