diff --git a/bench/x16-transpose.cc b/bench/x16-transpose.cc
index d219f779e632..9c6edb093d48 100644
--- a/bench/x16-transpose.cc
+++ b/bench/x16-transpose.cc
@@ -62,6 +62,9 @@ static void BenchmarkKernelSize(benchmark::internal::Benchmark* b)
   b->Args({32, 32});
   b->Args({64, 64});
   b->Args({117, 117});
+  b->Args({128, 128});
+  b->Args({256, 256});
+  b->Args({512, 512});
   b->Args({1024, 1024});
 }
 
diff --git a/bench/x32-transpose.cc b/bench/x32-transpose.cc
index b61fb584ad4c..927d56d092a3 100644
--- a/bench/x32-transpose.cc
+++ b/bench/x32-transpose.cc
@@ -62,6 +62,8 @@ static void BenchmarkKernelSize(benchmark::internal::Benchmark* b)
   b->Args({32, 32});
   b->Args({64, 64});
   b->Args({128, 128});
+  b->Args({256, 256});
+  b->Args({512, 512});
   b->Args({1024, 1024});
 }
 
diff --git a/bench/x64-transpose.cc b/bench/x64-transpose.cc
index 72674965a25b..20cf040f2012 100644
--- a/bench/x64-transpose.cc
+++ b/bench/x64-transpose.cc
@@ -62,6 +62,9 @@ static void BenchmarkKernelSize(benchmark::internal::Benchmark* b)
   b->Args({32, 32});
   b->Args({64, 64});
   b->Args({117, 117});
+  b->Args({128, 128});
+  b->Args({256, 256});
+  b->Args({512, 512});
   b->Args({1024, 1024});
 }
 
diff --git a/bench/x8-transpose.cc b/bench/x8-transpose.cc
index f05a8b825e7d..268c3c0625ee 100644
--- a/bench/x8-transpose.cc
+++ b/bench/x8-transpose.cc
@@ -62,6 +62,9 @@ static void BenchmarkKernelSize(benchmark::internal::Benchmark* b)
   b->Args({32, 32});
   b->Args({64, 64});
   b->Args({117, 117});
+  b->Args({128, 128});
+  b->Args({256, 256});
+  b->Args({512, 512});
   b->Args({1024, 1024});
 }