diff --git a/backend/src/packages/chaiNNer_ncnn/ncnn/processing/upscale_image.py b/backend/src/packages/chaiNNer_ncnn/ncnn/processing/upscale_image.py
index 046ca786d8..e947eb3cf7 100644
--- a/backend/src/packages/chaiNNer_ncnn/ncnn/processing/upscale_image.py
+++ b/backend/src/packages/chaiNNer_ncnn/ncnn/processing/upscale_image.py
@@ -153,8 +153,9 @@ def estimate_cpu():
     name="Upscale Image",
     description=(
         "Upscale an image with NCNN. Unlike PyTorch, NCNN has GPU support on all"
-        " devices, assuming your drivers support Vulkan. Select a manual"
-        " number of tiles if you are having issues with the automatic mode."
+        " devices, assuming your drivers support Vulkan. Select a lower memory limit or"
+        " manually adjust the number of tiles if you are experiencing issues with the"
+        " automatic mode."
     ),
     icon="NCNN",
     inputs=[
@@ -170,9 +171,10 @@ def estimate_cpu():
             "Generally it's recommended to use the largest tile size possible for best"
             " performance (with the ideal scenario being no tiling at all), but"
             " depending on the model and image size, this may not be possible.",
-            "If you are having issues with the automatic mode, you can manually select"
-            " a tile size. On certain machines, a very small tile size such as 256 or"
-            " 128 might be required for it to work at all.",
+            "If you are having issues with the automatic mode, you can either set a"
+            " lower memory limit or manually select a tile size. On certain machines, a"
+            " very small tile size such as 256 or 128 might be required for it to work"
+            " at all.",
         ),
     ],
     outputs=[
diff --git a/backend/src/packages/chaiNNer_pytorch/pytorch/processing/upscale_image.py b/backend/src/packages/chaiNNer_pytorch/pytorch/processing/upscale_image.py
index 5dc63af469..a4750b40b0 100644
--- a/backend/src/packages/chaiNNer_pytorch/pytorch/processing/upscale_image.py
+++ b/backend/src/packages/chaiNNer_pytorch/pytorch/processing/upscale_image.py
@@ -102,8 +102,9 @@ def estimate():
     schema_id="chainner:pytorch:upscale_image",
     name="Upscale Image",
     description=(
-        "Upscales an image using a PyTorch Super-Resolution model. Select a manual"
-        " number of tiles if you are having issues with the automatic mode."
+        "Upscales an image using a PyTorch Super-Resolution model. Select a lower"
+        " memory limit or manually adjust the number of tiles if you are experiencing"
+        " issues with the automatic mode."
     ),
     icon="PyTorch",
     inputs=[
@@ -124,9 +125,10 @@ def estimate():
                 "Generally it's recommended to use the largest tile size possible for"
                 " best performance (with the ideal scenario being no tiling at all),"
                 " but depending on the model and image size, this may not be possible.",
-                "If you are having issues with the automatic mode, you can manually"
-                " select a tile size. Sometimes, a manually selected tile size may be"
-                " faster than what the automatic mode picks.",
+                "If you are having issues with the automatic mode, you can either"
+                " select a lower memory limit or manually select a tile size."
+                " Sometimes, a manually selected tile size may be faster than what the"
+                " automatic mode picks.",
                 hint=True,
             )
         ),