diff --git a/fbgemm_gpu/experimental/gen_ai/src/quantize/cutlass_extensions.cu b/fbgemm_gpu/experimental/gen_ai/src/quantize/cutlass_extensions.cu
index d973d28ac..ef44aedb1 100644
--- a/fbgemm_gpu/experimental/gen_ai/src/quantize/cutlass_extensions.cu
+++ b/fbgemm_gpu/experimental/gen_ai/src/quantize/cutlass_extensions.cu
@@ -1896,6 +1896,14 @@ at::Tensor f8f8bf16_cublas(
       "CUDA version is older than 12.0"); // requires CUDA>=12
 }
 at::Tensor f8f8bf16(
+    at::Tensor XQ, // FP8
+    at::Tensor WQ, // FP8
+    at::Tensor scale,
+    bool use_fast_accum) {
+  throw std::runtime_error(
+      "CUDA version is older than 12.0"); // requires CUDA>=12
+}
+at::Tensor f8f8bf16_tensorwise(
     at::Tensor XQ, // FP8
     at::Tensor WQ, // FP8
     double scale,