diff --git a/cmake/CMakeLists.txt b/cmake/CMakeLists.txt
index 8184f9784d..a6956f5f5d 100644
--- a/cmake/CMakeLists.txt
+++ b/cmake/CMakeLists.txt
@@ -527,11 +527,6 @@ foreach(PKG_WITH_INCL KSPACE PYTHON ML-IAP VORONOI COLVARS ML-HDNNP MDI MOLFILE
   endif()
 endforeach()
 
-# AMOEBA is not compatible with single precision FFTs
-if(PKG_AMOEBA AND FFT_SINGLE)
-  message(FATAL_ERROR "Package AMOEBA is not compatible with single precision FFTs")
-endif()
-
 # optionally enable building script wrappers using swig
 option(WITH_SWIG "Build scripting language wrappers with SWIG" OFF)
 if(WITH_SWIG)
diff --git a/cmake/Modules/Packages/GPU.cmake b/cmake/Modules/Packages/GPU.cmake
index 89e15e548b..2c766a2540 100644
--- a/cmake/Modules/Packages/GPU.cmake
+++ b/cmake/Modules/Packages/GPU.cmake
@@ -31,6 +31,10 @@ endif()
 option(GPU_DEBUG "Enable debugging code of the GPU package" OFF)
 mark_as_advanced(GPU_DEBUG)
 
+if(PKG_AMOEBA AND FFT_SINGLE)
+  message(FATAL_ERROR "GPU acceleration of AMOEBA is not (yet) compatible with single precision FFT")
+endif()
+
 file(GLOB GPU_LIB_SOURCES ${CONFIGURE_DEPENDS} ${LAMMPS_LIB_SOURCE_DIR}/gpu/[^.]*.cpp)
 file(MAKE_DIRECTORY ${LAMMPS_LIB_BINARY_DIR}/gpu)
 
diff --git a/doc/src/pair_amoeba.rst b/doc/src/pair_amoeba.rst
index 79b3daf22f..6ef92a6938 100644
--- a/doc/src/pair_amoeba.rst
+++ b/doc/src/pair_amoeba.rst
@@ -206,6 +206,9 @@ These pair styles can only be used via the *pair* keyword of the
   when compiling the GPU package for OpenCL has a few known issues
   when running on integrated GPUs and the calculation may crash.
 
+  The GPU accelerated pair styles are also not (yet) compatible
+  with single precision FFTs.
+
 ----------
 
 Restrictions