GPU Support by Release
To use your GPU with MATLAB®, you must install a recent graphics driver. Best practice is to ensure you
have the latest driver for your device. Installing the driver is sufficient for most
uses of GPUs in MATLAB, including gpuArray and GPU-enabled MATLAB functions. You can download the latest drivers for your GPU device at
NVIDIA Driver Downloads.
Supported GPUs
To see support for NVIDIA® GPU architectures by MATLAB release, consult the following table.
– Built-in binary support. 
– Supported by NVIDIA's forward compatibility (requires
recompilation). The MATLAB release was built before this GPU architecture was available.
The CUDA® driver must recompile the GPU libraries because your device is
more recent than the libraries. The first time you access the GPU from
MATLAB, the compilation can take several minutes. Increase the
CUDA cache size to prevent a recurrence of this delay. For
instructions, see Increase the CUDA Cache Size.
– Support for Kepler and Maxwell GPU
architectures will be removed in a future release. At that time, using a GPU
with MATLAB will require a GPU device with compute capability 6.0 or
greater. In R2020a, Kepler and Maxwell GPUs are still supported. MATLAB generates a warning the first time you use a Kepler or Maxwell
GPU.
The cc numbers show the compute capability of the GPU
architecture. To check your GPU compute capability, see
ComputeCapability in the output of the gpuDevice function. Alternatively,
see CUDA GPUs
(NVIDIA).
| MATLAB Release | Turing (cc7.5) | Volta (cc7.0, cc7.2) | Pascal (cc6.x) | Maxwell (cc5.x) | Kepler (cc3.x) | Fermi (cc2.x) | Tesla (cc1.3) | CUDA Toolkit Version |
|---|---|---|---|---|---|---|---|---|
| R2020a |
|
|
|
|
| 10.1 | ||
| R2019b |
|
|
|
|
| 10.1 | ||
| R2019a |
|
|
|
|
| 10.0 | ||
| R2018b |
|
|
|
|
| 9.1 | ||
| R2018a |
|
|
|
|
| 9.0 | ||
| R2017b |
|
|
|
|
|
| 8.0 | |
| R2017a |
|
|
|
|
|
| 8.0 | |
| R2016b |
|
|
|
|
|
| 7.5 | |
| R2016a |
|
|
|
|
|
| 7.5 | |
| R2015b |
|
|
|
|
|
| 7.0 | |
| R2015a |
|
|
|
|
|
| 6.5 | |
| R2014b |
|
|
|
|
|
| 6.0 | |
| R2014a |
|
|
|
|
|
|
| 5.5 |
| R2013b |
|
|
|
|
|
|
| 5.0 |
| R2013a |
|
|
|
|
|
|
| 5.0 |
| R2012b |
|
|
|
|
|
|
| 4.2 |
| R2012a |
|
|
|
|
|
|
| 4.0 |
| R2011b |
|
|
|
|
|
|
| 4.0 |
| R2011a |
|
|
|
|
|
|
| 3.2 |
| R2010b |
|
|
|
|
|
|
| 3.1 |
CUDA Toolkit
If you want to use CUDAKernel objects or use GPU Coder, you must install a CUDA Toolkit. The CUDA Toolkit contains CUDA libraries and tools for compilation.
| Task | Requirements |
|---|---|
Use | Get the latest graphics driver at NVIDIA Driver Downloads. You do not need the CUDA Toolkit as well. |
Create and use CUDAKernel objects or use GPU Coder. | Install the version of the CUDA Toolkit supported by your MATLAB release. |
For more information about generating CUDA code in MATLAB, see Run MEX-Functions Containing CUDA Code and Run CUDA or PTX Code on GPU. Not all compilers supported by the CUDA Toolkit are supported in MATLAB.
For more information about the CUDA Toolkit and to download your supported version, see CUDA Toolkit Archive (NVIDIA).
Increase the CUDA Cache Size
If your GPU architecture does not have built-in binary support in your MATLAB release, the graphics driver must compile and cache the GPU libraries.
This process can take a few minutes the first time you access the GPU from
MATLAB. To increase the CUDA cache size to prevent a recurrence of this delay, set the environment
variable CUDA_CACHE_MAXSIZE to a minimum of
536870912 (512 MB). For help setting an environment variable,
see this example: Set the MATLABPATH Environment Variable (MATLAB).