diff --git a/Makefile.config.example b/Makefile.config.example
index 24ca632783a..de73707b020 100644
--- a/Makefile.config.example
+++ b/Makefile.config.example
@@ -33,18 +33,21 @@ CUDA_DIR := /usr/local/cuda
 # CUDA_DIR := /usr
 
 # CUDA architecture setting: going with all of them.
-# For CUDA < 6.0, comment the *_50 through *_61 lines for compatibility.
-# For CUDA < 8.0, comment the *_60 and *_61 lines for compatibility.
-# For CUDA >= 9.0, comment the *_20 and *_21 lines for compatibility.
-CUDA_ARCH := -gencode arch=compute_20,code=sm_20 \
-		-gencode arch=compute_20,code=sm_21 \
+# For CUDA < 6.0, comment all the lines starting from *_50 for compatibility.
+# For CUDA < 8.0, comment all the lines starting from *_60 for compatibility.
+CUDA_ARCH := \
 		-gencode arch=compute_30,code=sm_30 \
 		-gencode arch=compute_35,code=sm_35 \
 		-gencode arch=compute_50,code=sm_50 \
 		-gencode arch=compute_52,code=sm_52 \
 		-gencode arch=compute_60,code=sm_60 \
 		-gencode arch=compute_61,code=sm_61 \
-		-gencode arch=compute_61,code=compute_61
+		-gencode arch=compute_62,code=sm_62 \
+		-gencode arch=compute_70,code=sm_70 \
+		-gencode arch=compute_72,code=sm_72 \
+		-gencode arch=compute_75,code=sm_75 \
+		-gencode arch=compute_80,code=sm_80 \
+		-gencode arch=compute_86,code=sm_86
 
 # BLAS choice:
 # atlas for ATLAS (default)
diff --git a/cmake/Cuda.cmake b/cmake/Cuda.cmake
index e03feabffcb..b5ca3477d00 100644
--- a/cmake/Cuda.cmake
+++ b/cmake/Cuda.cmake
@@ -4,7 +4,20 @@ endif()
 
 # Known NVIDIA GPU achitectures Caffe can be compiled for.
 # This list will be used for CUDA_ARCH_NAME = All option
-set(Caffe_known_gpu_archs "20 21(20) 30 35 50 60 61")
+# Kepler (CUDA >= 5)
+set(KEPLER "30 35 37")
+# Maxwell (CUDA >= 6)
+set(MAXWELL "50 52 53")
+# Pascal (CUDA >= 8)
+set(PASCAL "60 61 62")
+# Volta (CUDA >= 9)
+set(VOLTA "70 72")
+# Turing (CUDA >= 10)
+set(TURING "75")
+# Ampere (CUDA >= 11)
+set(AMPERE "80 86")
+
+set(Caffe_known_gpu_archs "${KEPLER} ${MAXWELL} ${PASCAL} ${VOLTA} ${TURING} ${AMPERE}")
 
 ################################################################################################
 # A function for automatic detection of GPUs installed  (if autodetection is enabled)
@@ -56,7 +69,7 @@ endfunction()
 #   caffe_select_nvcc_arch_flags(out_variable)
 function(caffe_select_nvcc_arch_flags out_variable)
   # List of arch names
-  set(__archs_names "Fermi" "Kepler" "Maxwell" "Pascal" "All" "Manual")
+  set(__archs_names "Kepler (CUDA >= 5)" "Maxwell (CUDA >= 6)" "Pascal (CUDA >= 8)" "Volta (CUDA >= 9)" "Turing (CUDA >= 10)" "Ampere (CUDA >= 11)" "All" "Manual")
   set(__archs_name_default "All")
   if(NOT CMAKE_CROSSCOMPILING)
     list(APPEND __archs_names "Auto")
@@ -83,14 +96,20 @@ function(caffe_select_nvcc_arch_flags out_variable)
     unset(CUDA_ARCH_PTX CACHE)
   endif()
 
-  if(${CUDA_ARCH_NAME} STREQUAL "Fermi")
-    set(__cuda_arch_bin "20 21(20)")
-  elseif(${CUDA_ARCH_NAME} STREQUAL "Kepler")
-    set(__cuda_arch_bin "30 35")
-  elseif(${CUDA_ARCH_NAME} STREQUAL "Maxwell")
-    set(__cuda_arch_bin "50")
-  elseif(${CUDA_ARCH_NAME} STREQUAL "Pascal")
-    set(__cuda_arch_bin "60 61")
+  if (${CUDA_ARCH_NAME} STREQUAL "Fermi (3.2 <= CUDA <= 8)" AND NOT WIN32)
+    set(__cuda_arch_bin ${FERMI})
+  elseif (${CUDA_ARCH_NAME} STREQUAL "Kepler (CUDA >= 5)")
+    set(__cuda_arch_bin ${KEPLER})
+  elseif (${CUDA_ARCH_NAME} STREQUAL "Maxwell (CUDA >= 6)")
+    set(__cuda_arch_bin ${MAXWELL})
+  elseif (${CUDA_ARCH_NAME} STREQUAL "Pascal (CUDA >= 8)")
+    set(__cuda_arch_bin ${PASCAL})
+  elseif (${CUDA_ARCH_NAME} STREQUAL "Volta (CUDA >= 9)")
+    set(__cuda_arch_bin ${VOLTA})
+  elseif (${CUDA_ARCH_NAME} STREQUAL "Turing (CUDA >= 10)")
+    set(__cuda_arch_bin ${TURING})
+  elseif (${CUDA_ARCH_NAME} STREQUAL "Ampere (CUDA >= 11)")
+    set(__cuda_arch_bin ${AMPERE})
   elseif(${CUDA_ARCH_NAME} STREQUAL "All")
     set(__cuda_arch_bin ${Caffe_known_gpu_archs})
   elseif(${CUDA_ARCH_NAME} STREQUAL "Auto")