diff --git a/ext/MatrixAlgebraKitAMDGPUExt/MatrixAlgebraKitAMDGPUExt.jl b/ext/MatrixAlgebraKitAMDGPUExt/MatrixAlgebraKitAMDGPUExt.jl
index 14d849e5e..add91008e 100644
--- a/ext/MatrixAlgebraKitAMDGPUExt/MatrixAlgebraKitAMDGPUExt.jl
+++ b/ext/MatrixAlgebraKitAMDGPUExt/MatrixAlgebraKitAMDGPUExt.jl
@@ -35,6 +35,13 @@ end
 
 MatrixAlgebraKit.supports_svd_full(::ROCSOLVER, f::Symbol) = f in (:qr_iteration, :jacobi, :divide_and_conquer, :bisection)
 
+# one kernel for the whole triangle instead of one per column
+function MatrixAlgebraKit.uppertriangular!(A::StridedROCMatrix)
+    m, n = size(A)
+    A .= ifelse.((1:m) .> transpose(1:n), zero(eltype(A)), A)
+    return A
+end
+
 # rocSOLVER's `gesvd*_batched` functions take the batch as an array
 # of device pointers, so a group of equally sized matrices
 # doesn't need a copy into a 3D ROCArray.
diff --git a/ext/MatrixAlgebraKitCUDAExt/MatrixAlgebraKitCUDAExt.jl b/ext/MatrixAlgebraKitCUDAExt/MatrixAlgebraKitCUDAExt.jl
index cbcc9fbe6..3da8a7817 100644
--- a/ext/MatrixAlgebraKitCUDAExt/MatrixAlgebraKitCUDAExt.jl
+++ b/ext/MatrixAlgebraKitCUDAExt/MatrixAlgebraKitCUDAExt.jl
@@ -42,6 +42,13 @@ end
 
 MatrixAlgebraKit.prefers_ungqr(::CUSOLVER) = true
 
+# one kernel for the whole triangle instead of one per column
+function MatrixAlgebraKit.uppertriangular!(A::StridedCuMatrix)
+    m, n = size(A)
+    A .= ifelse.((1:m) .> transpose(1:n), zero(eltype(A)), A)
+    return A
+end
+
 MatrixAlgebraKit.supports_svd_full(::CUSOLVER, f::Symbol) = f in (:qr_iteration, :jacobi, :svd_polar)
 
 # `cusolverDnXgesvdjBatched` only accepts matrices up to 32x32