diff --git a/arm_compute/core/ITensorInfo.h b/arm_compute/core/ITensorInfo.h
index 0171e31..bc3a6be 100644
--- a/arm_compute/core/ITensorInfo.h
+++ b/arm_compute/core/ITensorInfo.h
@@ -240,6 +240,11 @@
      * @return True if its dynamic else false
      */
     virtual bool is_dynamic() const = 0;
+    /** Flag indicating whether the values of the tensor are constant, meaning that they can change on kernel/function execution.
+     *
+     * @return True if values are constant else false
+     */
+    virtual bool are_values_constant() const = 0;
     /** Set the flag whether the tensor size can be changed.
      *
      * @param[in] is_resizable Flag that marks the tensor if it can be changed or not.
@@ -247,6 +252,13 @@
      * @return Reference to this ITensorInfo object
      */
     virtual ITensorInfo &set_is_resizable(bool is_resizable) = 0;
+    /** Set the flag whether the tensor values can change during kernel/function execution.
+     *
+     * @param[in] are_values_constant Flag that marks the tensor values if they can be changed or not.
+     *
+     * @return Reference to this ITensorInfo object
+     */
+    virtual ITensorInfo &set_are_values_constant(bool are_values_constant) = 0;
     /** Valid region of the tensor. All elements in the valid region have defined values, i.e. are not undefined.
      *
      * @return The valid region.
diff --git a/arm_compute/core/SubTensorInfo.h b/arm_compute/core/SubTensorInfo.h
index 1b2278d..54836d0 100644
--- a/arm_compute/core/SubTensorInfo.h
+++ b/arm_compute/core/SubTensorInfo.h
@@ -196,12 +196,23 @@
         ARM_COMPUTE_ERROR_ON(_parent == nullptr);
         return _parent->is_dynamic();
     }
+    bool are_values_constant() const override
+    {
+        ARM_COMPUTE_ERROR_ON(_parent == nullptr);
+        return _parent->are_values_constant();
+    }
     ITensorInfo &set_is_resizable(bool is_resizable) override
     {
         ARM_COMPUTE_ERROR_ON(_parent == nullptr);
         _parent->set_is_resizable(is_resizable);
         return *this;
     }
+    ITensorInfo &set_are_values_constant(bool are_values_constant) override
+    {
+        ARM_COMPUTE_ERROR_ON(_parent == nullptr);
+        _parent->set_are_values_constant(are_values_constant);
+        return *this;
+    }
     ValidRegion valid_region() const override
     {
         return _valid_region;
diff --git a/arm_compute/core/TensorInfo.h b/arm_compute/core/TensorInfo.h
index a433084..9bc8680 100644
--- a/arm_compute/core/TensorInfo.h
+++ b/arm_compute/core/TensorInfo.h
@@ -267,6 +267,10 @@
     {
         return std::find(std::cbegin(_dims_state), std::cend(_dims_state), get_dynamic_state_value()) != std::cend(_dims_state);
     }
+    bool are_values_constant() const override
+    {
+        return _are_values_constant;
+    }
     ITensorInfo &set_is_resizable(bool is_resizable) override
     {
         _is_resizable = is_resizable;
@@ -288,6 +292,11 @@
     {
         return _data_layout;
     }
+    ITensorInfo &set_are_values_constant(bool are_values_constant) override
+    {
+        _are_values_constant = are_values_constant;
+        return *this;
+    }
 
 private:
     /** Calculates strides, offset and total size resulting from the specified padding around the XY plane.
@@ -309,6 +318,7 @@
     PaddingSize      _padding;
     QuantizationInfo _quantization_info;
     DataLayout       _data_layout;
+    bool             _are_values_constant;
 };
 } // namespace arm_compute
 #endif /*ARM_COMPUTE_TENSORINFO_H */
diff --git a/arm_compute/core/Types.h b/arm_compute/core/Types.h
index 9c00cbc..36b77b8 100644
--- a/arm_compute/core/Types.h
+++ b/arm_compute/core/Types.h
@@ -1544,7 +1544,6 @@
     bool       transpose_weights{ true };                  /**<  Transpose weights if true. */
     bool       are_weights_reshaped{ false };              /**<  Reshape the weights tensor if false. */
     bool       retain_internal_weights{ false };           /**<  Retain internal reshaped weights. */
-    bool       constant_weights{ true };                   /**<  If false, weights can vary between runs. */
     /* Other parameters */
     bool fp_mixed_precision{ false }; /**<  Use wider accumulators (32 bit instead of 16 for FP16) to improve accuracy. */
 
@@ -1951,9 +1950,8 @@
           _fast_math(false),
           _fp_mixed_precision(false),
           _broadcast_bias(false),
-          _pretranpose_B(true),
-          _activation_info(),
-          _constant_weights(true)
+          _pretranspose_B(true),
+          _activation_info()
     {
     }
     /** Constructor
@@ -1971,11 +1969,10 @@
      * @param[in] fast_math                   (Optional) Use a data type of shorter width to improve performance
      * @param[in] broadcast_bias              (Optional) Broadcast the shape of the bias tensor from a vector to a matrix.
      * @param[in] activation_info             (Optional) Activation to apply after the matrix multiplication
-     * @param[in] constant_weights            (Optional) Weights have constant values throughout multiple executions
      */
     GEMMInfo(bool is_a_reshaped, bool is_b_reshaped, bool reshape_b_only_on_first_run, int depth_output_gemm3d = 0, bool reinterpret_input_as_3d = false, bool retain_internal_weights = false,
              GEMMLowpOutputStageInfo gemmlowp_output_stage = GEMMLowpOutputStageInfo(), bool fp_mixed_precision = false, bool fast_math = false, bool broadcast_bias = false,
-             const ActivationLayerInfo &activation_info = ActivationLayerInfo(), bool constant_weights = true) noexcept
+             const ActivationLayerInfo &activation_info = ActivationLayerInfo()) noexcept
         : _is_a_reshaped(is_a_reshaped),
           _is_b_reshaped(is_b_reshaped),
           _reshape_b_only_on_first_run(reshape_b_only_on_first_run),
@@ -1986,9 +1983,8 @@
           _fast_math(fast_math),
           _fp_mixed_precision(fp_mixed_precision),
           _broadcast_bias(broadcast_bias),
-          _pretranpose_B(reshape_b_only_on_first_run),
-          _activation_info(activation_info),
-          _constant_weights(constant_weights)
+          _pretranspose_B(reshape_b_only_on_first_run),
+          _activation_info(activation_info)
     {
     }
     /** Flag which specifies if the matrix A has been reshaped
@@ -2085,17 +2081,17 @@
      *
      * @return True if b should be pre-transposed else false.
      */
-    bool pretranpose_B() const
+    bool pretranspose_B() const
     {
-        return _pretranpose_B;
+        return _pretranspose_B;
     };
     /** Set pre-transpose b flag
      *
      * @param[in] flag Flag to set
      */
-    void set_pretranpose_B(bool flag)
+    void set_pretranspose_B(bool flag)
     {
-        _pretranpose_B = flag;
+        _pretranspose_B = flag;
     }
     /** Activation layer to apply after the matrix multiplication
      *
@@ -2113,14 +2109,6 @@
     {
         _activation_info = activation_info;
     }
-    /** Flag which specifies if the values of the weights tensor are constant throughout multiple executions or not
-     *
-     * @return True if the weights tensor is constant
-     */
-    bool constant_weights() const
-    {
-        return _constant_weights;
-    };
 
 private:
     bool                    _is_a_reshaped;
@@ -2133,9 +2121,8 @@
     bool                    _fast_math;
     bool                    _fp_mixed_precision;
     bool                    _broadcast_bias;
-    bool                    _pretranpose_B;
+    bool                    _pretranspose_B;
     ActivationLayerInfo     _activation_info;
-    bool                    _constant_weights;
 };
 
 /** Winograd information */