diff --git a/docs/source/en/api/attnprocessor.md b/docs/source/en/api/attnprocessor.md
index 7a4812e0961e..0b11c1f5bc5d 100644
--- a/docs/source/en/api/attnprocessor.md
+++ b/docs/source/en/api/attnprocessor.md
@@ -17,6 +17,9 @@ An attention processor is a class for applying different types of attention mech
 ## CustomDiffusionAttnProcessor
 [[autodoc]] models.attention_processor.CustomDiffusionAttnProcessor
 
+## CustomDiffusionAttnProcessor2_0
+[[autodoc]] models.attention_processor.CustomDiffusionAttnProcessor2_0
+
 ## AttnAddedKVProcessor
 [[autodoc]] models.attention_processor.AttnAddedKVProcessor
 
@@ -39,4 +42,4 @@ An attention processor is a class for applying different types of attention mech
 [[autodoc]] models.attention_processor.SlicedAttnProcessor
 
 ## SlicedAttnAddedKVProcessor
-[[autodoc]] models.attention_processor.SlicedAttnAddedKVProcessor
\ No newline at end of file
+[[autodoc]] models.attention_processor.SlicedAttnAddedKVProcessor
diff --git a/src/diffusers/models/attention_processor.py b/src/diffusers/models/attention_processor.py
index ea226836932e..46e9b92ca91c 100644
--- a/src/diffusers/models/attention_processor.py
+++ b/src/diffusers/models/attention_processor.py
@@ -1468,7 +1468,8 @@ def __call__(self, attn: Attention, hidden_states, encoder_hidden_states=None, a
 
 class CustomDiffusionAttnProcessor2_0(nn.Module):
     r"""
-    Processor for implementing attention for the Custom Diffusion method.
+    Processor for implementing attention for the Custom Diffusion method
+    using PyTorch 2.0’s memory-efficient scaled dot-product attention.
 
     Args:
         train_kv (`bool`, defaults to `True`):
@@ -1758,6 +1759,7 @@ def __call__(self, attn: "Attention", hidden_states, encoder_hidden_states=None,
     LoRAAttnAddedKVProcessor,
     CustomDiffusionAttnProcessor,
     CustomDiffusionXFormersAttnProcessor,
+    CustomDiffusionAttnProcessor2_0,
 ]
 
 LORA_ATTENTION_PROCESSORS = (