diff --git a/ext/MatrixAlgebraKitAMDGPUExt/MatrixAlgebraKitAMDGPUExt.jl b/ext/MatrixAlgebraKitAMDGPUExt/MatrixAlgebraKitAMDGPUExt.jl
index 976c1dc04..43b6601f1 100644
--- a/ext/MatrixAlgebraKitAMDGPUExt/MatrixAlgebraKitAMDGPUExt.jl
+++ b/ext/MatrixAlgebraKitAMDGPUExt/MatrixAlgebraKitAMDGPUExt.jl
@@ -41,10 +41,14 @@ MatrixAlgebraKit.supports_svd_full(::ROCSOLVER, f::Symbol) = f in (:qr_iteration
 # 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.
-MatrixAlgebraKit.supports_pointer_batch(::AbstractAlgorithm, ::ROCSOLVER, ::Type{<:StridedROCMatrix{<:BlasFloat}}) = true
-MatrixAlgebraKit.supports_ragged_batch(::typeof(svd_full!), ::AbstractAlgorithm, ::ROCSOLVER, ::Type{<:AnyROCArray}) = true
-MatrixAlgebraKit.supports_ragged_batch(::typeof(svd_compact!), ::AbstractAlgorithm, ::ROCSOLVER, ::Type{<:AnyROCArray}) = true
-MatrixAlgebraKit.supports_ragged_batch(::typeof(svd_vals!), ::AbstractAlgorithm, ::ROCSOLVER, ::Type{<:AnyROCArray}) = true
+# opt-in per algorithm to avoid picking up `DiagonalAlgorithm`
+# which doesn't support batching
+for alg in (:Bisection, :Jacobi, :QRIteration, :DivideAndConquer)
+    @eval MatrixAlgebraKit.supports_pointer_batch(::$alg, ::ROCSOLVER, ::Type{<:StridedROCMatrix{<:BlasFloat}}) = true
+    @eval MatrixAlgebraKit.supports_ragged_batch(::typeof(svd_full!), ::$alg, ::ROCSOLVER, ::Type{<:AnyROCArray}) = true
+    @eval MatrixAlgebraKit.supports_ragged_batch(::typeof(svd_compact!), ::$alg, ::ROCSOLVER, ::Type{<:AnyROCArray}) = true
+    @eval MatrixAlgebraKit.supports_ragged_batch(::typeof(svd_vals!), ::$alg, ::ROCSOLVER, ::Type{<:AnyROCArray}) = true
+end
 
 function gesvd!(::ROCSOLVER, A::StridedROCMatrix, S::StridedROCVector, U::StridedROCMatrix, Vᴴ::StridedROCMatrix; kwargs...)
     m, n = size(A)
diff --git a/test/decompositions/svd.jl b/test/decompositions/svd.jl
index b7a4a6ef3..8c0c671f8 100644
--- a/test/decompositions/svd.jl
+++ b/test/decompositions/svd.jl
@@ -68,6 +68,7 @@ if CUDA.functional()
         TestSuite.test_svd_algs_batched_oversized(CuMatrix{T}, (Jacobi(),), batch_size)
         TestSuite.test_svd_algs_batched_ragged_support(CuMatrix{T}, (Jacobi(),), batch_size)
         TestSuite.test_svd_algs_batched_ragged_support(CuMatrix{T}, (QRIteration(),), batch_size; supported = false)
+        TestSuite.test_svd_algs_batched_ragged_support(Diagonal{T, CuVector{T}}, (DiagonalAlgorithm(),), batch_size; supported = false)
     end
 
     # Randomized SVD:
@@ -104,6 +105,7 @@ if AMDGPU.functional()
         TestSuite.seed_rng!(123)
         AMD_SVD_ALGS = (QRIteration(), Jacobi(), DivideAndConquer(), Bisection())
         TestSuite.test_svd_algs_batched_ragged_support(ROCMatrix{T}, AMD_SVD_ALGS, batch_size)
+        TestSuite.test_svd_algs_batched_ragged_support(Diagonal{T, ROCVector{T}}, (DiagonalAlgorithm(),), batch_size; supported = false)
     end
 
     @testset "Bisection with min(m, n) == 1 $(testargs_summary(T, sz))" for T in BLASFloats,