diff --git a/onnxruntime/core/providers/cuda/cuda_execution_provider.cc b/onnxruntime/core/providers/cuda/cuda_execution_provider.cc
index 6b83987403..b33be3c7e0 100644
--- a/onnxruntime/core/providers/cuda/cuda_execution_provider.cc
+++ b/onnxruntime/core/providers/cuda/cuda_execution_provider.cc
@@ -300,9 +300,9 @@ class ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain,
 class ONNX_OPERATOR_VERSIONED_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 7, 8, float, Greater);
 class ONNX_OPERATOR_VERSIONED_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 7, 8, double, Greater);
 class ONNX_OPERATOR_VERSIONED_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 7, 8, MLFloat16, Greater);
-class ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 7, bool, Equal);
-class ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 7, int32_t, Equal);
-class ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 7, int64_t, Equal);
+class ONNX_OPERATOR_VERSIONED_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 7, 10, bool, Equal);
+class ONNX_OPERATOR_VERSIONED_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 7, 10, int32_t, Equal);
+class ONNX_OPERATOR_VERSIONED_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 7, 10, int64_t, Equal);
 class ONNX_OPERATOR_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 8, Expand);
 class ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 9, int32_t, Greater);
 class ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 9, int64_t, Greater);
@@ -653,6 +653,9 @@ class ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain,
 class ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 11, double, MaxPool);
 class ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 11, MLFloat16, MaxPool);
 class ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 11, float, Clip);
+class ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 11, bool, Equal);
+class ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 11, int32_t, Equal);
+class ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 11, int64_t, Equal);
 
 static void RegisterCudaKernels(KernelRegistry& kernel_registry) {
   static const BuildKernelCreateInfoFn function_table[] = {
@@ -749,9 +752,9 @@ static void RegisterCudaKernels(KernelRegistry& kernel_registry) {
       BuildKernelCreateInfo<ONNX_OPERATOR_VERSIONED_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 7, 8, float, Greater)>,
       BuildKernelCreateInfo<ONNX_OPERATOR_VERSIONED_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 7, 8, double, Greater)>,
       BuildKernelCreateInfo<ONNX_OPERATOR_VERSIONED_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 7, 8, MLFloat16, Greater)>,
-      BuildKernelCreateInfo<ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 7, bool, Equal)>,
-      BuildKernelCreateInfo<ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 7, int32_t, Equal)>,
-      BuildKernelCreateInfo<ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 7, int64_t, Equal)>,
+      BuildKernelCreateInfo<ONNX_OPERATOR_VERSIONED_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 7, 10, bool, Equal)>,
+      BuildKernelCreateInfo<ONNX_OPERATOR_VERSIONED_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 7, 10, int32_t, Equal)>,
+      BuildKernelCreateInfo<ONNX_OPERATOR_VERSIONED_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 7, 10, int64_t, Equal)>,
       BuildKernelCreateInfo<ONNX_OPERATOR_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 8, Expand)>,
       BuildKernelCreateInfo<ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 9, int32_t, Greater)>,
       BuildKernelCreateInfo<ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 9, int64_t, Greater)>,
@@ -1101,6 +1104,9 @@ static void RegisterCudaKernels(KernelRegistry& kernel_registry) {
       BuildKernelCreateInfo<ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 11, double, MaxPool)>,
       BuildKernelCreateInfo<ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 11, MLFloat16, MaxPool)>,      
       BuildKernelCreateInfo<ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 11, float, Clip)>,
+      BuildKernelCreateInfo<ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 11, bool, Equal)>,
+      BuildKernelCreateInfo<ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 11, int32_t, Equal)>,
+      BuildKernelCreateInfo<ONNX_OPERATOR_TYPED_KERNEL_CLASS_NAME(kCudaExecutionProvider, kOnnxDomain, 11, int64_t, Equal)>,
   };
 
   for (auto& function_table_entry : function_table) {
diff --git a/onnxruntime/core/providers/cuda/math/binary_elementwise_ops.cc b/onnxruntime/core/providers/cuda/math/binary_elementwise_ops.cc
index f2941ab5e2..23865e28e5 100644
--- a/onnxruntime/core/providers/cuda/math/binary_elementwise_ops.cc
+++ b/onnxruntime/core/providers/cuda/math/binary_elementwise_ops.cc
@@ -176,6 +176,11 @@ Status BinaryElementwise<ShouldBroadcast>::Prepare(OpKernelContext* context, Bin
   BINARY_ELEMENTWISE_REGISTER_KERNEL_TYPED(name, ver, int32_t) \
   BINARY_ELEMENTWISE_REGISTER_KERNEL_TYPED(name, ver, int64_t)
 
+#define BINARY_OP_REGISTER_VERSIONED_OIL(name, startver, endver)                      \
+  BINARY_ELEMENTWISE_REGISTER_KERNEL_VERSIONED_TYPED(name, startver, endver, bool)    \
+  BINARY_ELEMENTWISE_REGISTER_KERNEL_VERSIONED_TYPED(name, startver, endver, int32_t) \
+  BINARY_ELEMENTWISE_REGISTER_KERNEL_VERSIONED_TYPED(name, startver, endver, int64_t)
+
 #define BINARY_LOGICALOP_REGISTER_OIL(name, ver)                         \
   BINARY_ELEMENTWISE_LOGICALOP_REGISTER_KERNEL_TYPED(name, ver, bool)    \
   BINARY_ELEMENTWISE_LOGICALOP_REGISTER_KERNEL_TYPED(name, ver, int32_t) \
@@ -379,7 +384,8 @@ Status Less<T>::ComputeInternal(OpKernelContext* context) const {
 BINARY_OP_REGISTER_UZILHFD(Sum, 8)
 BINARY_OP_REGISTER_VERSIONED_UZILHFD(Sum, 6, 7)
 BINARY_LOGICALOP_REGISTER_UZILHFD(Greater, 9)
-BINARY_OP_REGISTER_OIL(Equal, 7)
+BINARY_OP_REGISTER_VERSIONED_OIL(Equal, 7, 10)
+BINARY_OP_REGISTER_OIL(Equal, 11)
 BINARY_OP_REGISTER_VERSIONED_HFD(Greater, 7, 8)
 BINARY_OP_REGISTER_HFD(Max, 8)
 BINARY_OP_REGISTER_VERSIONED_HFD(Max, 6, 7)