We are pleased to announce Anaconda 2.0, the newest version of our enterprise-ready Python distribution. Available for free on Windows, Mac OS X and Linux, Anaconda includes almost 200 of the most popular numerical and scientific Python libraries used by scientists, engineers and data analysts, with an integrated and flexible installer.

Version 2.0 reduces the size of the installers, adds installers for Python 3.4, updates over 40 packages, and moves from PySide to PyQt. Highlights include:

  • The addition of Anaconda 3 installers, which are based on Python 3.4. This means that you can now set up Python 3 directly. (Previously, this required creating an additional conda environment or using Miniconda 3.)
  • A reduction in the size of the current installers. The Anaconda installers were becoming too big, due to the inclusion of some large packages. These packages have been removed from the installers, but are still supported and available in the repository, meaning you can still use conda to install them. These packages include mayavi, netcdf4, vtk, and 10 others.
  • The move from PySide to PyQt. PyQt is now the default Qt binding, as some users were experiencing stability problems with spyder.

A comprehensive, detailed list of changes to Anaconda in version 2.0 can be found in the change log.

To download Anaconda, please visit our download page. For help installing Anaconda, please visit our documentation page.

If you currently have Anaconda 1.9.2 (or a previous version) installed, update to Anaconda 2.0 with:

conda update conda
conda update anaconda

If you are interested in long-term support for Anaconda packages, check out Anaconda Server. Anaconda Server provides you with an internal version of the current Anaconda repository that can cache & manage packages from PyPI (and R packages from CRAN), as well as support for all previous versions of libraries in the repository. It’s a great way for teams to synchronize their use of open source libraries, and streamline the production deployment process.

Additionally, if you are interested in training, virtual training, or consulting help on your data science, data integration, or scientific computing problems, please contact us at [email protected]. For general questions about Anaconda, please email the public Anaconda support mailing list.