diff --git a/cub/cub/device/dispatch/dispatch_segmented_sort.cuh b/cub/cub/device/dispatch/dispatch_segmented_sort.cuh
index af085f65274..ef912463d6d 100644
--- a/cub/cub/device/dispatch/dispatch_segmented_sort.cuh
+++ b/cub/cub/device/dispatch/dispatch_segmented_sort.cuh
@@ -60,6 +60,8 @@
 #include <thrust/system/cuda/detail/core/triple_chevron_launch.h>
 
 #include <cuda/cmath>
+#include <cuda/std/__algorithm/max.h>
+#include <cuda/std/__algorithm/min.h>
 #include <cuda/std/type_traits>
 
 #include <type_traits>
@@ -964,7 +966,7 @@ struct DispatchSegmentedSort
         constexpr auto num_segments_per_invocation_limit =
           static_cast<global_segment_offset_t>(::cuda::std::numeric_limits<int>::max());
         auto const max_num_segments_per_invocation = static_cast<global_segment_offset_t>(
-          ::cuda::std::min(static_cast<global_segment_offset_t>(num_segments), num_segments_per_invocation_limit));
+          (::cuda::std::min)(static_cast<global_segment_offset_t>(num_segments), num_segments_per_invocation_limit));
 
         large_and_medium_segments_indices.grow(max_num_segments_per_invocation);
         small_segments_indices.grow(max_num_segments_per_invocation);
diff --git a/cub/cub/util_arch.cuh b/cub/cub/util_arch.cuh
index 3c6aea5cc5c..a2093ae288b 100644
--- a/cub/cub/util_arch.cuh
+++ b/cub/cub/util_arch.cuh
@@ -121,20 +121,20 @@ template <int Nominal4ByteBlockThreads, int Nominal4ByteItemsPerThread, typename
 struct RegBoundScaling
 {
   static constexpr int ITEMS_PER_THREAD =
-    ::cuda::std::max(1, Nominal4ByteItemsPerThread * 4 / ::cuda::std::max(4, int{sizeof(T)}));
-  static constexpr int BLOCK_THREADS =
-    ::cuda::std::min(Nominal4ByteBlockThreads,
-                     ::cuda::ceil_div(int{detail::max_smem_per_block} / (int{sizeof(T)} * ITEMS_PER_THREAD), 32) * 32);
+    (::cuda::std::max)(1, Nominal4ByteItemsPerThread * 4 / (::cuda::std::max)(4, int{sizeof(T)}));
+  static constexpr int BLOCK_THREADS = (::cuda::std::min)(
+    Nominal4ByteBlockThreads,
+    ::cuda::ceil_div(int{detail::max_smem_per_block} / (int{sizeof(T)} * ITEMS_PER_THREAD), 32) * 32);
 };
 
 template <int Nominal4ByteBlockThreads, int Nominal4ByteItemsPerThread, typename T>
 struct MemBoundScaling
 {
-  static constexpr int ITEMS_PER_THREAD = ::cuda::std::max(
-    1, ::cuda::std::min(Nominal4ByteItemsPerThread * 4 / int{sizeof(T)}, Nominal4ByteItemsPerThread * 2));
-  static constexpr int BLOCK_THREADS =
-    ::cuda::std::min(Nominal4ByteBlockThreads,
-                     ::cuda::ceil_div(int{detail::max_smem_per_block} / (int{sizeof(T)} * ITEMS_PER_THREAD), 32) * 32);
+  static constexpr int ITEMS_PER_THREAD = (::cuda::std::max)(
+    1, (::cuda::std::min)(Nominal4ByteItemsPerThread * 4 / int{sizeof(T)}, Nominal4ByteItemsPerThread * 2));
+  static constexpr int BLOCK_THREADS = (::cuda::std::min)(
+    Nominal4ByteBlockThreads,
+    ::cuda::ceil_div(int{detail::max_smem_per_block} / (int{sizeof(T)} * ITEMS_PER_THREAD), 32) * 32);
 };
 
 #endif // Do not document
diff --git a/cub/cub/util_macro.cuh b/cub/cub/util_macro.cuh
index e95a7136f43..3314123c9ab 100644
--- a/cub/cub/util_macro.cuh
+++ b/cub/cub/util_macro.cuh
@@ -49,6 +49,26 @@
 
 CUB_NAMESPACE_BEGIN
 
+#ifndef _CCCL_DOXYGEN_INVOKED // Do not document
+#  define CUB_PREVENT_MACRO_SUBSTITUTION
+template <typename T, typename U>
+CCCL_DEPRECATED_BECAUSE("Use cuda::std::min from <cuda/std/functional> instead")
+constexpr _CCCL_HOST_DEVICE auto min CUB_PREVENT_MACRO_SUBSTITUTION(T&& t, U&& u)
+  -> decltype(t < u ? ::cuda::std::forward<T>(t) : ::cuda::std::forward<U>(u))
+{
+  return t < u ? ::cuda::std::forward<T>(t) : ::cuda::std::forward<U>(u);
+}
+
+template <typename T, typename U>
+CCCL_DEPRECATED_BECAUSE("Use cuda::std::max from <cuda/std/functional> instead")
+constexpr _CCCL_HOST_DEVICE auto max CUB_PREVENT_MACRO_SUBSTITUTION(T&& t, U&& u)
+  -> decltype(t < u ? ::cuda::std::forward<U>(u) : ::cuda::std::forward<T>(t))
+{
+  return t < u ? ::cuda::std::forward<U>(u) : ::cuda::std::forward<T>(t);
+}
+#  undef CUB_PREVENT_MACRO_SUBSTITUTION
+#endif
+
 #ifndef CUB_MAX
 /// Select maximum(a, b)
 /// Deprecated since [2.8]
diff --git a/cub/test/catch2_radix_sort_helper.cuh b/cub/test/catch2_radix_sort_helper.cuh
index 9b7870f6ef0..642b2aed4f1 100644
--- a/cub/test/catch2_radix_sort_helper.cuh
+++ b/cub/test/catch2_radix_sort_helper.cuh
@@ -69,7 +69,7 @@ struct segment_iterator
 
   __host__ __device__ OffsetT operator()(std::int64_t x) const
   {
-    return ::cuda::std::min(last, x * Step);
+    return (::cuda::std::min)(last, x * Step);
   }
 };