diff --git a/extensions-builtin/sdnext-modernui b/extensions-builtin/sdnext-modernui
index 357985697..4edd7fc8b 160000
--- a/extensions-builtin/sdnext-modernui
+++ b/extensions-builtin/sdnext-modernui
@@ -1 +1 @@
-Subproject commit 357985697de0a457e401bccbc41132322648d0ba
+Subproject commit 4edd7fc8beb87488d98eb2f1b405a1b95c9163df
diff --git a/modules/caption/joytag.py b/modules/caption/joytag.py
index 346e296f1..97a05b8b5 100644
--- a/modules/caption/joytag.py
+++ b/modules/caption/joytag.py
@@ -14,10 +14,10 @@ import torch.nn as nn
 import torch.nn.functional as F
 from transformers.activations import QuickGELUActivation
 import torchvision
+import torchvision.transforms.functional as TVF
 import einops
 from einops.layers.torch import Rearrange
 import huggingface_hub
-from modules import images_sharpfin
 from modules import shared, devices, sd_models
 
 
@@ -1035,8 +1035,8 @@ def prepare_image(image: Image.Image, target_size: int) -> torch.Tensor:
     padded_image.paste(image, (pad_left, pad_top))
     if max_dim != target_size:
         padded_image = padded_image.resize((target_size, target_size), Image.Resampling.LANCZOS)
-    image_tensor = images_sharpfin.to_tensor(padded_image)
-    image_tensor = images_sharpfin.normalize(image_tensor, mean=[0.48145466, 0.4578275, 0.40821073], std=[0.26862954, 0.26130258, 0.27577711])
+    image_tensor = TVF.pil_to_tensor(padded_image) / 255.0
+    image_tensor = TVF.normalize(image_tensor, mean=[0.48145466, 0.4578275, 0.40821073], std=[0.26862954, 0.26130258, 0.27577711])
     return image_tensor