diff --git a/CMakeLists.txt b/CMakeLists.txt
index 231dc9bab..7e9db1ac5 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -741,6 +741,7 @@ if(CUDA_FOUND)
                        gpu_resources
                        gpu_smart_pointers
                        is_gpu_pointer
+                       throw_macros_on_gpu
                        PROPERTIES LABELS ${gpu_label})
 endif()
 
diff --git a/CMakeLists_files.cmake b/CMakeLists_files.cmake
index 28d3fefa9..7cb01265e 100644
--- a/CMakeLists_files.cmake
+++ b/CMakeLists_files.cmake
@@ -484,6 +484,8 @@ if (HAVE_CUDA)
   ADD_CUDA_OR_HIP_FILE(TEST_SOURCE_FILES tests test_gpu_smart_pointers.cu)
   ADD_CUDA_OR_HIP_FILE(TEST_SOURCE_FILES tests test_gpu_resources.cu)
   ADD_CUDA_OR_HIP_FILE(TEST_SOURCE_FILES tests test_is_gpu_pointer.cpp)
+  ADD_CUDA_OR_HIP_FILE(TEST_SOURCE_FILES tests test_throw_macros_on_gpu.cu)
+
 
   # for loop providing the flag --expt-relaxed-constexpr to fix some cuda issues with constexpr
   if(NOT CONVERT_CUDA_TO_HIP)
diff --git a/tests/gpuistl/test_throw_macros_on_gpu.cu b/tests/gpuistl/test_throw_macros_on_gpu.cu
new file mode 100644
index 000000000..800ed6149
--- /dev/null
+++ b/tests/gpuistl/test_throw_macros_on_gpu.cu
@@ -0,0 +1,109 @@
+/*
+  Copyright 2025 Equinor ASA
+
+  This file is part of the Open Porous Media project (OPM).
+  OPM is free software: you can redistribute it and/or modify
+  it under the terms of the GNU General Public License as published by
+  the Free Software Foundation, either version 3 of the License, or
+  (at your option) any later version.
+  OPM is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+  You should have received a copy of the GNU General Public License
+  along with OPM.  If not, see <http://www.gnu.org/licenses/>.
+*/
+#include <boost/test/tools/old/interface.hpp>
+#include <config.h>
+#include <stdexcept>
+
+#define BOOST_TEST_MODULE TestThrowMacrosOnGPU
+
+#include <cuda.h>
+#include <cuda_runtime.h>
+#include <boost/test/unit_test.hpp>
+#include <opm/common/ErrorMacros.hpp>
+#include <opm/simulators/linalg/gpuistl/detail/gpu_safe_call.hpp>
+
+namespace {
+__global__ void codeThatContainsMacros(bool call) {
+    if (call) {
+        OPM_THROW(std::logic_error, "Something went wrong");
+        OPM_THROW_NOLOG(std::logic_error, "Something went wrong");
+        OPM_THROW_PROBLEM(std::logic_error, "Something went wrong");
+    }
+    OPM_ERROR_IF(!call, "Something went horribly wrong");
+}
+
+// TODO: Check if this is better on HIP
+#if 0 // I am leaving this here to show that this is not possible due to limitations in CUDA
+      // the assert will indeed cause an error, but the CUDA context will be broken for
+      // the rest of the lifetime of the process, see 
+      // https://forums.developer.nvidia.com/t/how-to-clear-cuda-errors/296393/5
+__global__ void checkThrow() {
+    OPM_THROW(std::logic_error, "Something went wrong");
+}
+
+__global__ void checkThrowNoLog() {
+    OPM_THROW_NOLOG(std::logic_error, "Something went wrong");
+}
+
+__global__ void checkThrowProblem() {
+    OPM_THROW_PROBLEM(std::logic_error, "Something went wrong");
+}
+
+__global__ void checkErrorIf() {
+    OPM_ERROR_IF(true, "Something went horribly wrong");
+}
+#endif
+}
+
+BOOST_AUTO_TEST_CASE(TestKernel)
+{
+    OPM_GPU_SAFE_CALL(cudaDeviceSynchronize());
+    OPM_GPU_SAFE_CALL(cudaGetLastError());
+    codeThatContainsMacros<<<1, 1>>>(false);
+    OPM_GPU_SAFE_CALL(cudaDeviceSynchronize());
+    OPM_GPU_SAFE_CALL(cudaGetLastError());
+
+    #if 0 // I am leaving this here to show that this is not possible due to limitations in CUDA
+          // the assert will indeed cause an error, but the CUDA context will be broken for
+          // the rest of the lifetime of the process, see 
+          // https://forums.developer.nvidia.com/t/how-to-clear-cuda-errors/296393/5
+    codeThatContainsMacros<<<1, 1>>>(true);
+    // Make sure this actually throws
+    BOOST_CHECK_THROW(OPM_GPU_SAFE_CALL(cudaDeviceSynchronize()), std::runtime_error);
+    OPM_GPU_SAFE_CALL(cudaDeviceReset());
+    OPM_GPU_SAFE_CALL(cudaGetLastError());
+
+    checkThrow<<<1, 1>>>();
+    BOOST_CHECK_THROW(OPM_GPU_SAFE_CALL(cudaDeviceSynchronize()), std::runtime_error);
+    OPM_GPU_SAFE_CALL(cudaDeviceReset());
+    OPM_GPU_SAFE_CALL(cudaGetLastError());
+
+    checkThrowNoLog<<<1, 1>>>();
+    BOOST_CHECK_THROW(OPM_GPU_SAFE_CALL(cudaDeviceSynchronize()), std::runtime_error);
+    OPM_GPU_SAFE_CALL(cudaDeviceReset());
+    OPM_GPU_SAFE_CALL(cudaGetLastError());
+
+    checkThrowProblem<<<1, 1>>>();
+    BOOST_CHECK_THROW(OPM_GPU_SAFE_CALL(cudaDeviceSynchronize()), std::runtime_error);
+    OPM_GPU_SAFE_CALL(cudaDeviceReset());
+    OPM_GPU_SAFE_CALL(cudaGetLastError());
+
+    checkErrorIf<<<1, 1>>>();
+    BOOST_CHECK_THROW(OPM_GPU_SAFE_CALL(cudaDeviceSynchronize()), std::runtime_error);
+    OPM_GPU_SAFE_CALL(cudaDeviceReset());
+    OPM_GPU_SAFE_CALL(cudaGetLastError());
+    #endif
+}
+
+BOOST_AUTO_TEST_CASE(TestOutsideKernel) 
+{
+    // This is to make sure that the macros work outside of kernels but inside a .cu file
+    // ie. inside a file compiled by nvcc/hipcc.
+    BOOST_CHECK_THROW(OPM_THROW(std::runtime_error, "THROW"), std::runtime_error);
+    BOOST_CHECK_THROW(OPM_THROW_NOLOG(std::runtime_error, "THROW_NOLOG"), std::runtime_error);
+    BOOST_CHECK_THROW(OPM_THROW_PROBLEM(std::runtime_error, "THROW_PROBLEM"), std::runtime_error);
+    BOOST_CHECK_THROW(OPM_ERROR_IF(true, "ERROR_IF"), std::logic_error);
+}