Environment variables

Here are the environment variables CuPy uses.


Path to the directory containing CUDA. The parent of the directory containing nvcc is used as default. When nvcc is not found, /usr/local/cuda is used. See Working with Custom CUDA Installation for details.


Path to the directory to store kernel cache. ${HOME}/.cupy/kernel_cache is used by default. See Overview for details.


If set to 1, CUDA source file will be saved along with compiled binary in the cache directory for debug purpose. It is disabled by default. Note: source file will not be saved if the compiled binary is already stored in the cache.


If set to 1, CUPY_CACHE_DIR (and its default) and CUPY_CACHE_SAVE_CUDA_SOURCE will be ignored, and the cache is in memory. This env var allows reducing disk I/O, but is ignoed when nvcc is set to be the compiler backend.


If set to 1, when CUDA kernel compilation fails, CuPy dumps CUDA kernel code to standard error. It is disabled by default.


If set to 1, CUDA kernel will be compiled with debug information (--device-debug and --generate-line-info). It is disabled by default.


The amount of memory that can be allocated for each device. The value can be specified in absolute bytes or fraction (e.g., "90%") of the total memory of each GPU. See Memory Management for details. 0 (unlimited) is used by default.


Set the seed for random number generators.


If set to 1, the following syntax is enabled: cupy_ndarray[:] = numpy_ndarray.


A comma-separated string of backend names (cub or cutensor) which indicates the acceleration backends used in CuPy operations and its priority. Default is empty string (all accelerators are disabled).


If set to 1, it allows CUDA libraries to use Tensor Cores TF32 compute for 32-bit floating point compute. The default is 0 and TF32 is not used.


This controls CuPy’s behavior as a Consumer. If set to 0, a stream synchronization will not be performed when a device array provided by an external library that implements the CUDA Array Interface is being consumed by CuPy. Default is 1. For more detail, see the Synchronization requirement in the CUDA Array Interface v3 documentation.


This controls CuPy’s behavior as a Producer. If set to 2, the CuPy stream on which the data is being operated will not be exported and thus the Consumer (another library) will not perform any stream synchronization. Default is 3. For more detail, see the Synchronization requirement in the CUDA Array Interface v3 documentation.


Define the compiler to use when compiling CUDA source. Note that most CuPy kernels are built with NVRTC; this environment is only effective for RawKernels/RawModules with nvcc backend or when using cub as the accelerator.

Moreover, as in any CUDA programs, all of the CUDA environment variables listed in the CUDA Toolkit Documentation will also be honored. When CUPY_ACCELERATORS or NVCC environment variables are set, g++-6 or later is required as the runtime host compiler. Please refer to Installing CuPy from Source for the details on how to install g++.

For installation

These environment variables are used during installation (building CuPy from source).


See the description above.


Path to the cuTENSOR root directory that contains lib and include directories. (experimental)


Define the compiler to use when compiling CUDA files.


Enforce CuPy to be installed against Python 3.5.0 (not recommended).


For building the ROCm support, see Building CuPy for ROCm for further detail.


To build CuPy for a particular CUDA architecture. For example, CUPY_NVCC_GENERATE_CODE="arch=compute_60,code=sm_60". For specifying multiple archs, concatenate the arch=... strings with semicolons (;). When this is not set, the default is to support all architectures.