diff --git a/js/web/docs/webnn-operators.md b/js/web/docs/webnn-operators.md
index 75652899b5e5e..8711d4d20e370 100644
--- a/js/web/docs/webnn-operators.md
+++ b/js/web/docs/webnn-operators.md
@@ -22,7 +22,7 @@ operators and the supported opset domain/versions in **WebNN EP** by ONNX Runtim
 | Clip | ai.onnx(7-10, 11, 12, 13+) | clamp | ✓ | ✓ | WebNN CPU backend only supports 3 specific ranges: [0.0, infinity], [-1.0, 1.0], [0.0, 6.0] (Chromium issue: https://issues.chromium.org/issues/326156496) |
 | Concat | ai.onnx(7-10, 11-12, 13+) | concat | ✓ | ✓ | |
 | Conv | ai.onnx(7-10, 11+) | conv2d | ✓ | ✓ | Only supports 3-D or 4-D input and 'W' (weight) |
-| ConvTranspose | ai.onnx(7-10, 11+) | convTranspose2d | ✗ | ✓ | Only supports 3-D or 4-D input and 'W' (weight). |
+| ConvTranspose | ai.onnx(7-10, 11+) | convTranspose2d | ✓ | ✓ | Only supports 3-D or 4-D input and 'W' (weight). WebNN CPU backend only supports default dilations and group |
 | Cos | ai.onnx(7+) | cos | ✓ | ✓ | |
 | Div | ai.onnx(7-12, 13, 14+) | div | ✓ | ✓ | |
 | Elu | ai.onnx(7+) | elu | ✓ | ✓ | WebNN CPU backend only supports 'alpha' value is 1.0 |
diff --git a/onnxruntime/core/providers/webnn/builders/helper.h b/onnxruntime/core/providers/webnn/builders/helper.h
index 496f886e5a076..63fd97abb9a9a 100644
--- a/onnxruntime/core/providers/webnn/builders/helper.h
+++ b/onnxruntime/core/providers/webnn/builders/helper.h
@@ -167,7 +167,7 @@ static const InlinedHashMap<std::string, WebnnOpInfo> op_map = {
     {"Concat", {"concat", true}},
     {"Conv", {"conv2d", true}},
     {"ConvInteger", {"conv2dInteger", false}},
-    {"ConvTranspose", {"convTranspose2d", false}},
+    {"ConvTranspose", {"convTranspose2d", true}},
     {"Cos", {"cos", true}},
     {"Div", {"div", true}},
     {"DequantizeLinear", {"dequantizeLinear", false}},
diff --git a/onnxruntime/core/providers/webnn/builders/impl/conv_op_builder.cc b/onnxruntime/core/providers/webnn/builders/impl/conv_op_builder.cc
index 4f3f7459a7b5b..22049d2519712 100644
--- a/onnxruntime/core/providers/webnn/builders/impl/conv_op_builder.cc
+++ b/onnxruntime/core/providers/webnn/builders/impl/conv_op_builder.cc
@@ -427,6 +427,22 @@ bool ConvOpBuilder::HasSupportedInputsImpl(const Node& node, const WebnnDeviceTy
     return false;
   }
 
+  // WebNN CPU backend (TFLite) only supports default dilations and group.
+  // https://source.chromium.org/chromium/chromium/src/+/main:services/webnn/tflite/graph_builder_tflite.cc;l=1040
+  if (device_type == WebnnDeviceType::CPU && op_type == "ConvTranspose") {
+    NodeAttrHelper helper(node);
+    const auto dilations = helper.Get("dilations", std::vector<int64_t>{1, 1});
+    const auto group = helper.Get("group", 1);
+    if (dilations[0] != 1 || (dilations.size() > 1 && dilations[1] != 1)) {
+      LOGS(logger, VERBOSE) << op_type << " for WebNN CPU backend only supports default dilation 1.";
+      return false;
+    }
+    if (group != 1) {
+      LOGS(logger, VERBOSE) << op_type << " for WebNN CPU backend only supports default group 1.";
+      return false;
+    }
+  }
+
   return true;
 }