diff --git a/modules/apg/pipeline_stable_diffusion_apg.py b/modules/apg/pipeline_stable_diffusion_apg.py
index 57aebafca..1508b3f13 100644
--- a/modules/apg/pipeline_stable_diffusion_apg.py
+++ b/modules/apg/pipeline_stable_diffusion_apg.py
@@ -404,7 +404,8 @@ class StableDiffusionPipelineAPG(
                 # representations. The `last_hidden_states` that we typically use for
                 # obtaining the final prompt representations passes through the LayerNorm
                 # layer.
-                prompt_embeds = self.text_encoder.text_model.final_layer_norm(prompt_embeds)
+                # transformers >=5.6 flattened CLIPTextModel; CLIPTextModelWithProjection still nests it under .text_model
+                prompt_embeds = getattr(self.text_encoder, 'text_model', self.text_encoder).final_layer_norm(prompt_embeds)
 
         if self.text_encoder is not None:
             prompt_embeds_dtype = self.text_encoder.dtype
diff --git a/modules/control/units/xs_pipe.py b/modules/control/units/xs_pipe.py
index 282b4f996..078293831 100644
--- a/modules/control/units/xs_pipe.py
+++ b/modules/control/units/xs_pipe.py
@@ -1321,7 +1321,8 @@ class StableDiffusionControlNetXSPipeline(
                 # representations. The `last_hidden_states` that we typically use for
                 # obtaining the final prompt representations passes through the LayerNorm
                 # layer.
-                prompt_embeds = self.text_encoder.text_model.final_layer_norm(prompt_embeds)
+                # transformers >=5.6 flattened CLIPTextModel; CLIPTextModelWithProjection still nests it under .text_model
+                prompt_embeds = getattr(self.text_encoder, 'text_model', self.text_encoder).final_layer_norm(prompt_embeds)
 
         if self.text_encoder is not None:
             prompt_embeds_dtype = self.text_encoder.dtype
diff --git a/modules/pag/pipe_sd.py b/modules/pag/pipe_sd.py
index 9a7af9bda..8e3f24429 100644
--- a/modules/pag/pipe_sd.py
+++ b/modules/pag/pipe_sd.py
@@ -619,7 +619,8 @@ class StableDiffusionPAGPipeline(
                 # representations. The `last_hidden_states` that we typically use for
                 # obtaining the final prompt representations passes through the LayerNorm
                 # layer.
-                prompt_embeds = self.text_encoder.text_model.final_layer_norm(prompt_embeds)
+                # transformers >=5.6 flattened CLIPTextModel; CLIPTextModelWithProjection still nests it under .text_model
+                prompt_embeds = getattr(self.text_encoder, 'text_model', self.text_encoder).final_layer_norm(prompt_embeds)
 
         if self.text_encoder is not None:
             prompt_embeds_dtype = self.text_encoder.dtype
diff --git a/scripts/differential_diffusion.py b/scripts/differential_diffusion.py
index 2843bf62f..11eb177af 100644
--- a/scripts/differential_diffusion.py
+++ b/scripts/differential_diffusion.py
@@ -1546,7 +1546,8 @@ class StableDiffusionDiffImg2ImgPipeline(DiffusionPipeline):
                 # representations. The `last_hidden_states` that we typically use for
                 # obtaining the final prompt representations passes through the LayerNorm
                 # layer.
-                prompt_embeds = self.text_encoder.text_model.final_layer_norm(prompt_embeds)
+                # transformers >=5.6 flattened CLIPTextModel; CLIPTextModelWithProjection still nests it under .text_model
+                prompt_embeds = getattr(self.text_encoder, 'text_model', self.text_encoder).final_layer_norm(prompt_embeds)
 
         if self.text_encoder is not None:
             prompt_embeds_dtype = self.text_encoder.dtype