diff --git a/ports/ncnn/fix_uwp.patch b/ports/ncnn/fix_uwp.patch
new file mode 100644
index 00000000000000..d678464aeaa3dc
--- /dev/null
+++ b/ports/ncnn/fix_uwp.patch
@@ -0,0 +1,42 @@
+diff --git a/src/cpu.cpp b/src/cpu.cpp
+index e56cd67..3ca1189 100644
+--- a/src/cpu.cpp
++++ b/src/cpu.cpp
+@@ -184,6 +184,13 @@ static int g_cpu_is_arm_a53_a55;
+ #endif // defined __ANDROID__ || defined __linux__
+ 
+ #if defined _WIN32
++#if WINAPI_FAMILY == WINAPI_FAMILY_APP
++static int detectisa(const void* /*some_inst*/)
++{
++    // uwp does not support seh  :(
++    return 0;
++}
++#else  // WINAPI_FAMILY == WINAPI_FAMILY_APP
+ static int g_sigill_caught = 0;
+ static jmp_buf g_jmpbuf;
+ 
+@@ -213,6 +220,7 @@ static int detectisa(const void* some_inst)
+ 
+     return g_sigill_caught ? 0 : 1;
+ }
++#endif // WINAPI_FAMILY == WINAPI_FAMILY_APP
+ 
+ #if defined(__i386__) || defined(__x86_64__) || defined(_M_IX86) || defined(_M_X64)
+ #ifdef _MSC_VER
+@@ -1966,12 +1974,12 @@ static void initialize_global_cpu_info()
+ 
+ #if defined _WIN32
+ #if __arm__
+-    g_cpu_support_arm_neon = detectisa(some_neon);
++    g_cpu_support_arm_neon = 1; // all modern windows arm devices have neon
+     g_cpu_support_arm_vfpv4 = detectisa(some_vfpv4);
+ #if __aarch64__
+     g_cpu_support_arm_cpuid = detectisa(some_cpuid);
+-    g_cpu_support_arm_asimdhp = detectisa(some_asimdhp);
+-    g_cpu_support_arm_asimddp = detectisa(some_asimddp);
++    g_cpu_support_arm_asimdhp = detectisa(some_asimdhp) || IsProcessorFeaturePresent(43); // dp implies hp
++    g_cpu_support_arm_asimddp = detectisa(some_asimddp) || IsProcessorFeaturePresent(43); // 43 is PF_ARM_V82_DP_INSTRUCTIONS_AVAILABLE
+     g_cpu_support_arm_asimdfhm = detectisa(some_asimdfhm);
+     g_cpu_support_arm_bf16 = detectisa(some_bf16);
+     g_cpu_support_arm_i8mm = detectisa(some_i8mm);
diff --git a/ports/ncnn/portfile.cmake b/ports/ncnn/portfile.cmake
index f8e846e23f2130..a3dec468a362ed 100644
--- a/ports/ncnn/portfile.cmake
+++ b/ports/ncnn/portfile.cmake
@@ -2,8 +2,10 @@ vcpkg_from_github(
     OUT_SOURCE_PATH SOURCE_PATH
     REPO Tencent/ncnn
     REF "${VERSION}"
-    SHA512 722966b3b30c5a4df81c6d45237b1821acc69db0c78350a41e3bc60e3f40c3dc64587ae0ab223635c468314c665e477ee7e0c2d3d4cccbc72bb15aeb56dcda6c
+    SHA512 31bc3c2f461a00241fb8f69ca6ea8cc590af6618856b1b84a048bde924e4b474fd883ad5d54dbfbdd1e5b59015889e15ffc4fbafccb3e42e052a02071f2017b1
     HEAD_REF master
+    PATCHES
+        fix_uwp.patch #https://github.com/Tencent/ncnn/pull/5328
 )
 
 string(COMPARE EQUAL "${VCPKG_LIBRARY_LINKAGE}" "dynamic" BUILD_SHARED)
diff --git a/ports/ncnn/vcpkg.json b/ports/ncnn/vcpkg.json
index fbf005e68ae177..28be929f74ac3d 100644
--- a/ports/ncnn/vcpkg.json
+++ b/ports/ncnn/vcpkg.json
@@ -1,6 +1,6 @@
 {
   "name": "ncnn",
-  "version": "20231027",
+  "version": "20240102",
   "description": "ncnn is a high-performance neural network inference computing framework.",
   "homepage": "https://github.com/Tencent/ncnn",
   "license": "BSD-3-Clause",
diff --git a/versions/baseline.json b/versions/baseline.json
index d20ea4f53b468d..090c416f9120bd 100644
--- a/versions/baseline.json
+++ b/versions/baseline.json
@@ -5905,7 +5905,7 @@
       "port-version": 2
     },
     "ncnn": {
-      "baseline": "20231027",
+      "baseline": "20240102",
       "port-version": 0
     },
     "ncurses": {
diff --git a/versions/n-/ncnn.json b/versions/n-/ncnn.json
index 348ed94bf90426..60b0168727164b 100644
--- a/versions/n-/ncnn.json
+++ b/versions/n-/ncnn.json
@@ -1,5 +1,10 @@
 {
   "versions": [
+    {
+      "git-tree": "322a0f4cfc94f9d73060a9636854a8e1260796fd",
+      "version": "20240102",
+      "port-version": 0
+    },
     {
       "git-tree": "2a6a2a9cc4c71cac799469f587478b987a63e0ed",
       "version": "20231027",