diff --git a/api/onnx_web/convert/diffusion/lora.py b/api/onnx_web/convert/diffusion/lora.py
index 29f51fc1..9fb16517 100644
--- a/api/onnx_web/convert/diffusion/lora.py
+++ b/api/onnx_web/convert/diffusion/lora.py
@@ -77,6 +77,13 @@ def fix_node_name(key: str):
         return fixed_name
 
 
+def kernel_slice(x: int, y: int, shape: Tuple[int, int, int, int]) -> Tuple[int, int]:
+    return (
+        max(x, shape[2]),
+        max(y, shape[3]),
+    )
+
+
 def blend_loras(
     _conversion: ServerContext,
     base_name: Union[str, ModelProto],
@@ -278,9 +285,12 @@ def blend_loras(
 
                         for w in range(kernel[0]):
                             for h in range(kernel[1]):
-                                weights[:, :, w, h] = up_weight.squeeze(3).squeeze(
+                                down_w, down_h = kernel_slice(w, h, down_weight.shape)
+                                up_w, up_h = kernel_slice(w, h, up_weight.shape)
+
+                                weights[:, :, w, h] = up_weight[:, :, up_w, up_h].squeeze(3).squeeze(
                                     2
-                                ) @ down_weight.squeeze(3).squeeze(2)
+                                ) @ down_weight[:, :, down_w, down_h].squeeze(3).squeeze(2)
 
                         np_weights = weights.numpy() * (alpha / dim)
                 else: