diff --git a/paddle/phi/api/yaml/legacy_api.yaml b/paddle/phi/api/yaml/legacy_api.yaml
index 165005e808ca9071d616ac4b637be58cca7730fb..6e87936e7298329fee39052cd4a186e79263956f 100755
--- a/paddle/phi/api/yaml/legacy_api.yaml
+++ b/paddle/phi/api/yaml/legacy_api.yaml
@@ -1798,15 +1798,6 @@
     func : mode
   backward : mode_grad
 
-- api : modulo
-  args : (Tensor x, Tensor y)
-  output : Tensor
-  infer_meta :
-    func : ElementwiseInferMeta
-  kernel :
-    func : modulo
-  backward : modulo_grad
-
 - api : momentum_
   args : (Tensor param, Tensor grad, Tensor velocity, Tensor learning_rate, Tensor master_param, float mu, bool use_nesterov = false, str regularization_method = "", float regularization_coeff = 0.0, bool multi_precision = false, float rescale_grad = 1.0f)
   output : Tensor(param_out), Tensor(velocity_out), Tensor(master_param_out)
@@ -2128,6 +2119,15 @@
     func : relu6
   backward : relu6_grad
 
+- api : remainder
+  args : (Tensor x, Tensor y)
+  output : Tensor
+  infer_meta :
+    func : ElementwiseInferMeta
+  kernel :
+    func : remainder
+  inplace : (x -> out)
+
 - api : renorm
   args : (Tensor x, float p, int axis, float max_norm)
   output : Tensor
diff --git a/paddle/phi/api/yaml/legacy_backward.yaml b/paddle/phi/api/yaml/legacy_backward.yaml
index 307e7c453a9e3aea472ad5f9ae3072018c51b5bf..24793465b53ab6ef06101b1d0f8bf444088fc0e6 100755
--- a/paddle/phi/api/yaml/legacy_backward.yaml
+++ b/paddle/phi/api/yaml/legacy_backward.yaml
@@ -1576,17 +1576,6 @@
   kernel :
     func : mode_grad
 
-- backward_api : modulo_grad
-  forward : modulo (Tensor x, Tensor y) -> Tensor(out)
-  args : (Tensor x, Tensor y, Tensor out_grad, int axis = -1)
-  output : Tensor(x_grad), Tensor(y_grad)
-  infer_meta :
-    func : GeneralBinaryGradInferMeta
-    param : [x, y]
-  kernel :
-    func : modulo_grad
-  no_need_buffer : x, y
-
 - backward_api : multi_dot_grad
   forward : multi_dot (Tensor[] x) -> Tensor(out)
   args : (Tensor[] x, Tensor out_grad)
diff --git a/paddle/phi/kernels/cpu/elementwise_kernel.cc b/paddle/phi/kernels/cpu/elementwise_kernel.cc
index f090ddd5bbe9a7bc77ddaf469913bed3fdbdce72..b7c3f3c84823f90377a31c1e57d302deadd6e944 100644
--- a/paddle/phi/kernels/cpu/elementwise_kernel.cc
+++ b/paddle/phi/kernels/cpu/elementwise_kernel.cc
@@ -46,21 +46,21 @@ void MinimumRawKernel(const Context& dev_ctx,
 }
 
 template <typename T, typename Context>
-void ModuloRawKernel(const Context& dev_ctx,
-                     const DenseTensor& x,
-                     const DenseTensor& y,
-                     int axis,
-                     DenseTensor* out) {
+void RemainderRawKernel(const Context& dev_ctx,
+                        const DenseTensor& x,
+                        const DenseTensor& y,
+                        int axis,
+                        DenseTensor* out) {
   // allocate memory for out
   dev_ctx.template Alloc<T>(out);
   auto x_dims = x.dims();
   auto y_dims = y.dims();
   if (x_dims.size() >= y_dims.size()) {
-    funcs::ElementwiseCompute<funcs::ModuloFunctor<T>, T>(
-        dev_ctx, x, y, axis, funcs::ModuloFunctor<T>(), out);
+    funcs::ElementwiseCompute<funcs::RemainderFunctor<T>, T>(
+        dev_ctx, x, y, axis, funcs::RemainderFunctor<T>(), out);
   } else {
-    funcs::ElementwiseCompute<funcs::InverseModuloFunctor<T>, T>(
-        dev_ctx, x, y, axis, funcs::InverseModuloFunctor<T>(), out);
+    funcs::ElementwiseCompute<funcs::InverseRemainderFunctor<T>, T>(
+        dev_ctx, x, y, axis, funcs::InverseRemainderFunctor<T>(), out);
   }
 }
 
@@ -139,10 +139,10 @@ PD_REGISTER_KERNEL(minimum_raw,
                    int,
                    int64_t,
                    phi::dtype::bfloat16) {}
-PD_REGISTER_KERNEL(modulo_raw,
+PD_REGISTER_KERNEL(remainder_raw,
                    CPU,
                    ALL_LAYOUT,
-                   phi::ModuloRawKernel,
+                   phi::RemainderRawKernel,
                    float,
                    double,
                    int,
diff --git a/paddle/phi/kernels/elementwise_kernel.cc b/paddle/phi/kernels/elementwise_kernel.cc
index 5e29eb5ace675fd2cf246827e585633b44cbb541..ebdb6299f352ff43ef0a940722e8599c4e56a308 100644
--- a/paddle/phi/kernels/elementwise_kernel.cc
+++ b/paddle/phi/kernels/elementwise_kernel.cc
@@ -38,12 +38,12 @@ void MinimumKernel(const Context& dev_ctx,
 }
 
 template <typename T, typename Context>
-void ModuloKernel(const Context& dev_ctx,
-                  const DenseTensor& x,
-                  const DenseTensor& y,
-                  DenseTensor* out) {
+void RemainderKernel(const Context& dev_ctx,
+                     const DenseTensor& x,
+                     const DenseTensor& y,
+                     DenseTensor* out) {
   int axis = -1;
-  ModuloRawKernel<T>(dev_ctx, x, y, axis, out);
+  RemainderRawKernel<T>(dev_ctx, x, y, axis, out);
 }
 
 template <typename T, typename Context>
@@ -96,8 +96,14 @@ PD_REGISTER_KERNEL(minimum,
                    int,
                    int64_t,
                    phi::dtype::bfloat16) {}
-PD_REGISTER_KERNEL(
-    modulo, CPU, ALL_LAYOUT, phi::ModuloKernel, float, double, int, int64_t) {}
+PD_REGISTER_KERNEL(remainder,
+                   CPU,
+                   ALL_LAYOUT,
+                   phi::RemainderKernel,
+                   float,
+                   double,
+                   int,
+                   int64_t) {}
 PD_REGISTER_KERNEL(
     floor_divide, CPU, ALL_LAYOUT, phi::FloorDivideKernel, int, int64_t) {}
 PD_REGISTER_KERNEL(elementwise_heaviside,
@@ -139,8 +145,14 @@ PD_REGISTER_KERNEL(minimum,
                    int64_t,
                    phi::dtype::float16,
                    phi::dtype::bfloat16) {}
-PD_REGISTER_KERNEL(
-    modulo, GPU, ALL_LAYOUT, phi::ModuloKernel, float, double, int, int64_t) {}
+PD_REGISTER_KERNEL(remainder,
+                   GPU,
+                   ALL_LAYOUT,
+                   phi::RemainderKernel,
+                   float,
+                   double,
+                   int,
+                   int64_t) {}
 PD_REGISTER_KERNEL(
     floor_divide, KPS, ALL_LAYOUT, phi::FloorDivideKernel, int, int64_t) {}
 PD_REGISTER_KERNEL(elementwise_heaviside,
diff --git a/paddle/phi/kernels/elementwise_kernel.h b/paddle/phi/kernels/elementwise_kernel.h
index a39da52e7e3b543f6ed43e89f728b7a9dec12a48..65040e1937a59a399549ae3637f1115e2d13bdec 100644
--- a/paddle/phi/kernels/elementwise_kernel.h
+++ b/paddle/phi/kernels/elementwise_kernel.h
@@ -60,18 +60,18 @@ void MinimumKernel(const Context& dev_ctx,
                    DenseTensor* out);
 
 template <typename T, typename Context>
-void ModuloRawKernel(const Context& dev_ctx,
+void RemainderRawKernel(const Context& dev_ctx,
+                        const DenseTensor& x,
+                        const DenseTensor& y,
+                        int axis,
+                        DenseTensor* out);
+
+template <typename T, typename Context>
+void RemainderKernel(const Context& dev_ctx,
                      const DenseTensor& x,
                      const DenseTensor& y,
-                     int axis,
                      DenseTensor* out);
 
-template <typename T, typename Context>
-void ModuloKernel(const Context& dev_ctx,
-                  const DenseTensor& x,
-                  const DenseTensor& y,
-                  DenseTensor* out);
-
 template <typename T, typename Context>
 void FloorDivideRawKernel(const Context& dev_ctx,
                           const DenseTensor& x,
@@ -134,13 +134,13 @@ DenseTensor Minimum(const Context& dev_ctx,
 }
 
 template <typename T, typename Context>
-DenseTensor Modulo(const Context& dev_ctx,
-                   const DenseTensor& x,
-                   const DenseTensor& y) {
+DenseTensor Remainder(const Context& dev_ctx,
+                      const DenseTensor& x,
+                      const DenseTensor& y) {
   DenseTensor dense_out;
   MetaTensor meta_out(&dense_out);
   ElementwiseInferMeta(x, y, &meta_out);
-  ModuloKernel<T, Context>(dev_ctx, x, y, &dense_out);
+  RemainderKernel<T, Context>(dev_ctx, x, y, &dense_out);
   return dense_out;
 }
 
diff --git a/paddle/phi/kernels/funcs/elementwise_functor.h b/paddle/phi/kernels/funcs/elementwise_functor.h
index e30ab8716b608942fcadf47b456271d9f1a8260f..bfbfd28abbe1941071ca1aa83e78422420a41b7c 100644
--- a/paddle/phi/kernels/funcs/elementwise_functor.h
+++ b/paddle/phi/kernels/funcs/elementwise_functor.h
@@ -21,7 +21,7 @@ limitations under the License. */
 #if defined(__xpu__)
 #include <xpu/runtime.h>
 
-#include "xpu/kernel/math_xpu2.h"  //pow()
+#include "xpu/kernel/math_xpu2.h"  // pow()
 #endif
 
 namespace phi {
@@ -499,7 +499,7 @@ struct MinGradXYFunctor {
 
 // Modulo
 template <typename T, typename Enable = void>
-struct ModuloFunctor {
+struct RemainderFunctor {
   inline HOSTDEVICE T operator()(const T a, const T b) const {
     T res = a % b;
 
@@ -511,7 +511,7 @@ struct ModuloFunctor {
 };
 
 template <typename T>
-struct ModuloFunctor<
+struct RemainderFunctor<
     T,
     typename std::enable_if_t<std::is_floating_point<T>::value>> {
   inline HOSTDEVICE T operator()(const T a, const T b) const {
@@ -525,7 +525,7 @@ struct ModuloFunctor<
 };
 
 template <typename T, typename Enable = void>
-struct InverseModuloFunctor {
+struct InverseRemainderFunctor {
   inline HOSTDEVICE T operator()(const T a, const T b) const {
     T res = b % a;
     if ((res != 0) && ((res < 0) != (a < 0))) res += a;
@@ -534,7 +534,7 @@ struct InverseModuloFunctor {
 };
 
 template <typename T>
-struct InverseModuloFunctor<
+struct InverseRemainderFunctor<
     T,
     typename std::enable_if_t<std::is_floating_point<T>::value>> {
   inline HOSTDEVICE T operator()(const T a, const T b) const {
diff --git a/paddle/phi/kernels/kps/elementwise_kernel.cu b/paddle/phi/kernels/kps/elementwise_kernel.cu
index d387096a70b75e0cced379b2a1093a484dc5be88..9da65c590b4d98c3d8fae639e81436f1b8840c1b 100644
--- a/paddle/phi/kernels/kps/elementwise_kernel.cu
+++ b/paddle/phi/kernels/kps/elementwise_kernel.cu
@@ -42,8 +42,8 @@ void MinimumKernel(const Context& dev_ctx,
   int axis = -1;
   MinimumRawKernel<T>(dev_ctx, x, y, axis, out);
 }
-// Create the definition of Modulo
-DEFINE_CUDA_ELEMENTWISE_OP(Modulo)
+// Create the definition of Remainder
+DEFINE_CUDA_ELEMENTWISE_OP(Remainder)
 // Create the definition of FloorDivide
 DEFINE_CUDA_ELEMENTWISE_OP(FloorDivide)
 template <typename T, typename Context>
@@ -118,10 +118,10 @@ PD_REGISTER_KERNEL(minimum_raw,
                    int64_t,
                    float16,
                    bfloat16) {}
-PD_REGISTER_KERNEL(modulo_raw,
+PD_REGISTER_KERNEL(remainder_raw,
                    KPS,
                    ALL_LAYOUT,
-                   phi::ModuloRawKernel,
+                   phi::RemainderRawKernel,
                    float,
                    double,
                    int,
diff --git a/paddle/phi/ops/compat/elementwise_sig.cc b/paddle/phi/ops/compat/elementwise_sig.cc
index 17fb1858373d9581e62ed89cd3e05abe744362fe..1d82ceaf1dea3b354e21a83af3daa8ca8ada7e7f 100644
--- a/paddle/phi/ops/compat/elementwise_sig.cc
+++ b/paddle/phi/ops/compat/elementwise_sig.cc
@@ -86,9 +86,9 @@ KernelSignature ElementwiseModOpArgumentMapping(
     const ArgumentMappingContext& ctx) {
   int axis = paddle::any_cast<int>(ctx.Attr("axis"));
   if (axis == -1) {
-    return KernelSignature("modulo", {"X", "Y"}, {}, {"Out"});
+    return KernelSignature("remainder", {"X", "Y"}, {}, {"Out"});
   }
-  return KernelSignature("modulo_raw", {"X", "Y"}, {"axis"}, {"Out"});
+  return KernelSignature("remainder_raw", {"X", "Y"}, {"axis"}, {"Out"});
 }
 
 KernelSignature ElementwiseFloorDivOpArgumentMapping(
@@ -247,7 +247,7 @@ PD_REGISTER_BASE_KERNEL_NAME(elementwise_mul, multiply);
 PD_REGISTER_BASE_KERNEL_NAME(elementwise_div, divide);
 PD_REGISTER_BASE_KERNEL_NAME(elementwise_max, maximum);
 PD_REGISTER_BASE_KERNEL_NAME(elementwise_min, minimum);
-PD_REGISTER_BASE_KERNEL_NAME(elementwise_mod, modulo);
+PD_REGISTER_BASE_KERNEL_NAME(elementwise_mod, remainder);
 PD_REGISTER_BASE_KERNEL_NAME(elementwise_floordiv, floor_divide);
 PD_REGISTER_BASE_KERNEL_NAME(elementwise_add_grad, add_grad);
 PD_REGISTER_BASE_KERNEL_NAME(elementwise_add_grad_grad, add_double_grad);
diff --git a/python/paddle/fluid/dygraph/math_op_patch.py b/python/paddle/fluid/dygraph/math_op_patch.py
index b66d1ee302231d7c6b9f8e4a1c6b5fd008133c47..b12c43fa17bdeca4db2516d3ead356306357d54f 100644
--- a/python/paddle/fluid/dygraph/math_op_patch.py
+++ b/python/paddle/fluid/dygraph/math_op_patch.py
@@ -395,7 +395,7 @@ def monkey_patch_math_varbase():
         if framework._in_eager_mode_ else
         ('__floordiv__',
          _binary_creator_('__floordiv__', 'elementwise_floordiv', False, None)),
-        ('__mod__', _binary_creator_('__mod__', 'modulo', False, None, True))
+        ('__mod__', _binary_creator_('__mod__', 'remainder', False, None, True))
         if framework._in_eager_mode_ else
         ('__mod__',
          _binary_creator_('__mod__', 'elementwise_mod', False, None)),
diff --git a/python/paddle/fluid/layers/nn.py b/python/paddle/fluid/layers/nn.py
index 48e6924660a468e91ca0039503a43279f6b0c9fb..ebb50604b7311f492d24e0b3582053c0ffa02096 100755
--- a/python/paddle/fluid/layers/nn.py
+++ b/python/paddle/fluid/layers/nn.py
@@ -205,7 +205,7 @@ OP_NAMEMAPPING = {
     'elementwise_sub': 'subtract',
     'elementwise_mul': 'multiply',
     'elementwise_div': 'divide',
-    'elementwise_mod': 'modulo',
+    'elementwise_mod': 'remainder',
 }
 
 
diff --git a/python/paddle/tensor/math.py b/python/paddle/tensor/math.py
index cd8bd888cc1fbdf3979cb729d614001114f86a41..0bad545025c1c2c0f3bda776eb670760d7f7837d 100644
--- a/python/paddle/tensor/math.py
+++ b/python/paddle/tensor/math.py
@@ -419,12 +419,11 @@ OP_NAMEMAPPING = {
     'elementwise_min': 'minimum',
     'elementwise_pow': 'elementwise_pow',
     'elementwise_floordiv': 'floor_divide',
-    'elementwise_mod': 'modulo',
     'elementwise_add': 'add',
     'elementwise_sub': 'subtract',
     'elementwise_mul': 'multiply',
     'elementwise_div': 'divide',
-    'elementwise_mod': 'modulo',
+    'elementwise_mod': 'remainder',
 }
 
 @dygraph_only