Welcome, guest | Sign In | My Account | Store | Cart

Notice! PyPM is being replaced with the ActiveState Platform, which enhances PyPM’s build and deploy capabilities. Create your free Platform account to download ActivePython or customize Python with the packages you require and get automatic updates.

Download
ActivePython

pycuda is unavailable in PyPM, because there aren't any builds for it in the package repositories. Click the linked icons to find out why.

 Python 2.7Python 3.2Python 3.3
Windows (32-bit)
2013.1.1Never BuiltWhy not?
2011.2.2 Failed View build log
2011.2.1 Failed View build log
2011.2 Failed View build log
2011.1.2 Failed View build log
2011.1 Failed View build log
0.94.2 Failed View build log
0.94.1 Failed View build log
0.94 Failed View build log
0.94rc Failed View build log
Windows (64-bit)
2013.1.1Never BuiltWhy not?
2011.2.2 Failed View build log
2011.2.1 Failed View build log
2011.2 Failed View build log
2011.1.2 Failed View build log
2011.1 Failed View build log
0.94.2 Failed View build log
0.94.1 Failed View build log
0.94 Failed View build log
0.94rc Failed View build log
Mac OS X (10.5+)
2013.1.1Never BuiltWhy not?
2011.2.2 Failed View build log
2011.2.1 Failed View build log
2011.2 Failed View build log
2011.1.2 Failed View build log
2011.1 Failed View build log
0.94.2 Failed View build log
0.94.1 Failed View build log
0.94 Failed View build log
0.94rc Failed View build log
Linux (32-bit)
2013.1.1 Failed View build log
2012.1 Failed View build log
2011.2.2 Failed View build log
2011.2.1 Failed View build log
2011.2 Failed View build log
2011.1.2 Failed View build log
2011.1 Failed View build log
0.94.2 Failed View build log
0.94.1 Failed View build log
0.94 Failed View build log
0.94rc Failed View build log
Linux (64-bit)
2013.1.1 Failed View build log
2012.1 Failed View build log
2011.2.2 Failed View build log
2011.2.1 Failed View build log
2011.2 Failed View build log
2011.1.2 Failed View build log
2011.1 Failed View build log
0.94.2 Failed View build log
0.94.1 Failed View build log
0.94 Failed View build log
0.94rc Failed View build log
2013.1.1 Failed View build log
 
Links
License
MIT
Depended by
Lastest release
version 2013.1.1 on Jul 13th, 2013

PyCUDA lets you access Nvidia's CUDA parallel computation API from Python. Several wrappers of the CUDA API already exist-so what's so special about PyCUDA?

  • Object cleanup tied to lifetime of objects. This idiom, often called RAII in C++, makes it much easier to write correct, leak- and crash-free code. PyCUDA knows about dependencies, too, so (for example) it won't detach from a context before all memory allocated in it is also freed.
  • Convenience. Abstractions like pycuda.driver.SourceModule and pycuda.gpuarray.GPUArray make CUDA programming even more convenient than with Nvidia's C-based runtime.
  • Completeness. PyCUDA puts the full power of CUDA's driver API at your disposal, if you wish. It also includes code for interoperability with OpenGL.
  • Automatic Error Checking. All CUDA errors are automatically translated into Python exceptions.
  • Speed. PyCUDA's base layer is written in C++, so all the niceties above are virtually free.
  • Helpful Documentation and a Wiki.

Relatedly, like-minded computing goodness for OpenCL is provided by PyCUDA's sister project PyOpenCL.

Subscribe to package updates

Last updated Jul 13th, 2013

What does the lock icon mean?

Builds marked with a lock icon are only available via PyPM to users with a current ActivePython Business Edition subscription.

Need custom builds or support?

ActivePython Enterprise Edition guarantees priority access to technical support, indemnification, expert consulting and quality-assured language builds.

Plan on re-distributing ActivePython?

Get re-distribution rights and eliminate legal risks with ActivePython OEM Edition.