iBOTS Learn: Environment Management with Pixi

Essential Computing Tools for Scientists

Environment Management with Pixi

Authors

Dr. Atle Rimehaug | Dr. Ole Bialas | Dr. Nicholas Del Grosso | Dr. Sangeetha Nandakumar

Download Materials

Most software we are using comes with a list of dependencies which, in turn, have their own dependencies. This means that any given piece of software requires a specific environment of other packages to function properly. The larger the tree of dependencies, the more likely it is that two pieces of software have a version conflict because they require different versions of the same dependency. This can be prevented by using virtual environments. Virtual environments are self-contained boxes that are used only within a specific narrow scope. By creating a dedicated virtual environment for every project we can isolate the software and avoid conflicts. We can also improve the reproducibility of our research because the virtual environment can be shared together with the code and data.

In this notebook, you are going to learn how to use Pixi - a fast and modern tool for managing virtual environments. Pixi allows us to set up new projects and install packages very quickly and with few steps. What’s more, Pixi keeps track of our environment as we add new dependencies. It creates a manifest called pixi.toml which contains a list of the dependencies we specified. It also creates a file called pixi.lock which contains a comprehensive description of the virtual environment (i.e. the whole tree of dependencies). These files allow us to exactly reproduce the computational environment.

Setup

The exercises in this notebook assume that you have installed Pixi and Visual Studio Code. If you haven’t, just click on the link and follow the instructions for your operating system on the website. To test if Pixi was installed correctly, open a terminal (in VSCode select View>Terminal from the menu bar or press Ctrl+`) and type pixi --version. If you are getting an error saying that the command was not found, try restarting VSCode.

pixi --version

pixi 0.58.0

Section 1: Creating and Managing Environments

Background

To create a new Pixi project, open the terminal and move to the directory where you want to initialize the workspace. On Windows, you can also use the file explorer to go to the target directory and Right Click > Open in Terminal. Once you are in the directory, run pixi init, which will create the manifest pixi.toml file containing all dependencies you specified. Once you start installing dependencies, Pixi will create a .pixi/ directory where the installed software is stored as well as a pixi.lock file that contains an exact description of the current computational environment. To activate the environment, type pixi shell. Once the environment is activated, you’ll have access to all of the software installed with Pixi.

Exercises

In the following exercises, you are going to create a new Pixi workspace, install Python, and activate the virtual environment. Here are the commands you need to know.

Code	Description
`cd my_dir/`	Change the directory to `my_dir/`
`cd ..`	Change the directory to the parent of the current directory
`mkdir new_dir/`	Create the directory `new_dir/`
`pixi init`	Initialize a new Pixi project with a `pixi.toml`
`pixi add python`	add python to your environment
`pixi add python==3.13`	add Python version 3.13 to your environment
`pixi remove python`	Remove python from your environment
`pixi shell`	Activate the virtual environment

Example: Create a new empty directory and move to that directory.

mkdir /tmp/new_project # create /tmp/new_project/
cd /tmp/new_project # move to /tmp/new_project/

/tmp/new_project

Exercise: Create another new directory and move there. You can use your file explorer or the mkdir and cd commands as shown in the example above.

Solution

mkdir /tmp/my_project  # create tmp/my_project/
cd /tmp/my_project  # move to tmp/my_project

/tmp/my_project

Exercise: Run pixi init to initialize a new workspace in the current directory and open the pixi.toml file in VSCode.

Solution

pixi init

✔ Created /tmp/my_project/pixi.toml

Exercise: Use pixi add to add Python to the environment.

Solution

pixi add python

✔ Added python >=3.14.4,<3.15

Exercise: Use pixi shell to activate the environment and run python --version to print the version of the current python installation.

NOTE: Typing only python opens a Python shell. You can exit by typing exit().

Solution

pixi shell
python --version

Python 3.14.4

Exercise: Use pixi remove to remove Python from the environment.

Solution

pixi remove python

✔ Removed python

Exercise: Add python==3.12 to the environment

Solution

pixi add python==3.12

✔ Added python==3.12

Exercise: Use pixi shell to activate the environment and run python --version to print the version of the current python installation.

Solution

pixi shell
python --version

Python 3.12.0

Section 2: Installing Python Packages with Pixi

Background

There are two prominent places for publishing Python packages: Conda-Forge and the Python Package Index (PyPI). A nice feature of Pixi is that it can resolve dependencies across both repositories (i.e. it can make sure there is no version conflict between Conda and PyPI packages).

Per default, Pixi installs packages from Conda but we can tell it to install PyPI packages by adding the --pypi flag. Conda-forge tests distribution and integration more systematically and is able bundle packages with non-Python dependencies (e.g. C-libraries). However, many libraries (especially smaller ones) are only available on PyPI. Thus, it is recommended that you install packages from Conda-Forge and only use PyPI if a given package is not available there.

Exercises

In the following exercises you are going to install Python packages from Conda-Forge and PyPI. Here are the commands you need to know:

Code	Description
`pixi add numpy`	Install `numpy` from Conda-Forge
`pixi add --pypi numpy`	Install package from PyPI
`pixi remove numpy`	Remove `numpy` from the environment
`pixi add matplotlib pandas`	Install multiple packages (`matplotlib` and `pandas`)
`pixi shell`	Activate the virtual environment
`pixi list`	View all currently installed packages

Exercise: Install the packages numpy and matplotlib from Conda-Forge.

Solution

pixi add numpy matplotlib

✔ Added numpy >=2.4.3,<3
✔ Added matplotlib >=3.10.9,<4

Exercise: Use pixi list to verify the packages were installed.

Solution

pixi list

Package                    Version       Build                 Size        Kind   Source
_openmp_mutex              4.5           20_gnu                28.3 KiB    conda  https://conda.anaconda.org/conda-forge/
alsa-lib                   1.2.15.3      hb03c661_0            571 KiB     conda  https://conda.anaconda.org/conda-forge/
brotli                     1.2.0         hed03a55_1            19.6 KiB    conda  https://conda.anaconda.org/conda-forge/
brotli-bin                 1.2.0         hb03c661_1            20.5 KiB    conda  https://conda.anaconda.org/conda-forge/
bzip2                      1.0.8         hda65f42_9            254.1 KiB   conda  https://conda.anaconda.org/conda-forge/
ca-certificates            2026.4.22     hbd8a1cb_0            128 KiB     conda  https://conda.anaconda.org/conda-forge/
cairo                      1.18.4        he90730b_1            966.3 KiB   conda  https://conda.anaconda.org/conda-forge/
contourpy                  1.3.3         py312h0a2e395_4       312.9 KiB   conda  https://conda.anaconda.org/conda-forge/
cycler                     0.12.1        pyhcf101f3_2          14.4 KiB    conda  https://conda.anaconda.org/conda-forge/
cyrus-sasl                 2.1.28        hac629b4_1            205.2 KiB   conda  https://conda.anaconda.org/conda-forge/
dbus                       1.16.2        h24cb091_1            437.2 KiB   conda  https://conda.anaconda.org/conda-forge/
double-conversion          3.4.0         hecca717_0            70.1 KiB    conda  https://conda.anaconda.org/conda-forge/
font-ttf-dejavu-sans-mono  2.37          hab24e00_0            388.1 KiB   conda  https://conda.anaconda.org/conda-forge/
font-ttf-inconsolata       3.000         h77eed37_0            94.3 KiB    conda  https://conda.anaconda.org/conda-forge/
font-ttf-source-code-pro   2.038         h77eed37_0            684.4 KiB   conda  https://conda.anaconda.org/conda-forge/
font-ttf-ubuntu            0.83          h77eed37_3            1.5 MiB     conda  https://conda.anaconda.org/conda-forge/
fontconfig                 2.17.1        h27c8c51_0            264.4 KiB   conda  https://conda.anaconda.org/conda-forge/
fonts-conda-ecosystem      1             0                     3.6 KiB     conda  https://conda.anaconda.org/conda-forge/
fonts-conda-forge          1             hc364b38_1            4 KiB       conda  https://conda.anaconda.org/conda-forge/
fonttools                  4.62.1        py312h8a5da7c_0       2.8 MiB     conda  https://conda.anaconda.org/conda-forge/
freetype                   2.14.3        ha770c72_0            169.8 KiB   conda  https://conda.anaconda.org/conda-forge/
graphite2                  1.3.14        hecca717_2            97.3 KiB    conda  https://conda.anaconda.org/conda-forge/
harfbuzz                   14.2.0        h6083320_0            2.2 MiB     conda  https://conda.anaconda.org/conda-forge/
icu                        78.3          h33c6efd_0            12.1 MiB    conda  https://conda.anaconda.org/conda-forge/
keyutils                   1.6.3         hb9d3cd8_0            130.9 KiB   conda  https://conda.anaconda.org/conda-forge/
kiwisolver                 1.5.0         py312h0a2e395_0       75.3 KiB    conda  https://conda.anaconda.org/conda-forge/
krb5                       1.22.2        ha1258a1_0            1.3 MiB     conda  https://conda.anaconda.org/conda-forge/
lcms2                      2.19          h0c24ade_0            244.7 KiB   conda  https://conda.anaconda.org/conda-forge/
ld_impl_linux-64           2.45.1        default_hbd61a6d_102  710.9 KiB   conda  https://conda.anaconda.org/conda-forge/
lerc                       4.1.0         hdb68285_0            255.4 KiB   conda  https://conda.anaconda.org/conda-forge/
libblas                    3.11.0        6_h4a7cf45_openblas   18.2 KiB    conda  https://conda.anaconda.org/conda-forge/
libbrotlicommon            1.2.0         hb03c661_1            78.1 KiB    conda  https://conda.anaconda.org/conda-forge/
libbrotlidec               1.2.0         hb03c661_1            33.8 KiB    conda  https://conda.anaconda.org/conda-forge/
libbrotlienc               1.2.0         hb03c661_1            291.4 KiB   conda  https://conda.anaconda.org/conda-forge/
libcblas                   3.11.0        6_h0358290_openblas   18.2 KiB    conda  https://conda.anaconda.org/conda-forge/
libclang13                 22.1.4        default_h746c552_0    12.2 MiB    conda  https://conda.anaconda.org/conda-forge/
libcups                    2.3.3         h7a8fb5f_6            4.3 MiB     conda  https://conda.anaconda.org/conda-forge/
libdeflate                 1.25          h17f619e_0            71.8 KiB    conda  https://conda.anaconda.org/conda-forge/
libdrm                     2.4.125       hb03c661_1            303.5 KiB   conda  https://conda.anaconda.org/conda-forge/
libedit                    3.1.20250104  pl5321h7949ede_0      131.5 KiB   conda  https://conda.anaconda.org/conda-forge/
libegl                     1.7.0         ha4b6fd6_2            43.8 KiB    conda  https://conda.anaconda.org/conda-forge/
libegl-devel               1.7.0         ha4b6fd6_2            29.7 KiB    conda  https://conda.anaconda.org/conda-forge/
libexpat                   2.7.5         hecca717_0            74.8 KiB    conda  https://conda.anaconda.org/conda-forge/
libffi                     3.5.2         h3435931_0            57.2 KiB    conda  https://conda.anaconda.org/conda-forge/
libfreetype                2.14.3        ha770c72_0            7.9 KiB     conda  https://conda.anaconda.org/conda-forge/
libfreetype6               2.14.3        h73754d4_0            375.6 KiB   conda  https://conda.anaconda.org/conda-forge/
libgcc                     15.2.0        he0feb66_18           1017.4 KiB  conda  https://conda.anaconda.org/conda-forge/
libgcc-ng                  15.2.0        h69a702a_18           26.9 KiB    conda  https://conda.anaconda.org/conda-forge/
libgfortran                15.2.0        h69a702a_18           26.9 KiB    conda  https://conda.anaconda.org/conda-forge/
libgfortran5               15.2.0        h68bc16d_18           2.4 MiB     conda  https://conda.anaconda.org/conda-forge/
libgl                      1.7.0         ha4b6fd6_2            131.6 KiB   conda  https://conda.anaconda.org/conda-forge/
libgl-devel                1.7.0         ha4b6fd6_2            111.2 KiB   conda  https://conda.anaconda.org/conda-forge/
libglib                    2.88.0        h6705ce6_0            4.6 MiB     conda  https://conda.anaconda.org/conda-forge/
libglvnd                   1.7.0         ha4b6fd6_2            129.4 KiB   conda  https://conda.anaconda.org/conda-forge/
libglx                     1.7.0         ha4b6fd6_2            73.7 KiB    conda  https://conda.anaconda.org/conda-forge/
libglx-devel               1.7.0         ha4b6fd6_2            25.8 KiB    conda  https://conda.anaconda.org/conda-forge/
libgomp                    15.2.0        he0feb66_18           589.1 KiB   conda  https://conda.anaconda.org/conda-forge/
libiconv                   1.18          h3b78370_2            771.7 KiB   conda  https://conda.anaconda.org/conda-forge/
libjpeg-turbo              3.1.4.1       hb03c661_0            619 KiB     conda  https://conda.anaconda.org/conda-forge/
liblapack                  3.11.0        6_h47877c9_openblas   18.2 KiB    conda  https://conda.anaconda.org/conda-forge/
libllvm22                  22.1.4        hf7376ad_0            42.2 MiB    conda  https://conda.anaconda.org/conda-forge/
liblzma                    5.8.3         hb03c661_0            110.8 KiB   conda  https://conda.anaconda.org/conda-forge/
liblzma-devel              5.8.3         hb03c661_0            479.9 KiB   conda  https://conda.anaconda.org/conda-forge/
libnsl                     2.0.1         hb9d3cd8_1            32.9 KiB    conda  https://conda.anaconda.org/conda-forge/
libntlm                    1.8           hb9d3cd8_0            32.6 KiB    conda  https://conda.anaconda.org/conda-forge/
libopenblas                0.3.32        pthreads_h94d23a6_0   5.7 MiB     conda  https://conda.anaconda.org/conda-forge/
libopengl                  1.7.0         ha4b6fd6_2            49.6 KiB    conda  https://conda.anaconda.org/conda-forge/
libpciaccess               0.18          hb9d3cd8_0            27.8 KiB    conda  https://conda.anaconda.org/conda-forge/
libpng                     1.6.58        h421ea60_0            310.3 KiB   conda  https://conda.anaconda.org/conda-forge/
libpq                      18.3          h9abb657_0            2.7 MiB     conda  https://conda.anaconda.org/conda-forge/
libsqlite                  3.53.0        hf4e2dac_0            936.4 KiB   conda  https://conda.anaconda.org/conda-forge/
libstdcxx                  15.2.0        h934c35e_18           5.6 MiB     conda  https://conda.anaconda.org/conda-forge/
libstdcxx-ng               15.2.0        hdf11a46_18           26.9 KiB    conda  https://conda.anaconda.org/conda-forge/
libtiff                    4.7.1         h9d88235_1            425.1 KiB   conda  https://conda.anaconda.org/conda-forge/
libuuid                    2.42          h5347b49_0            39.4 KiB    conda  https://conda.anaconda.org/conda-forge/
libvulkan-loader           1.4.341.0     h5279c79_0            195.1 KiB   conda  https://conda.anaconda.org/conda-forge/
libwebp-base               1.6.0         hd42ef1d_0            419 KiB     conda  https://conda.anaconda.org/conda-forge/
libxcb                     1.17.0        h8a09558_0            386.6 KiB   conda  https://conda.anaconda.org/conda-forge/
libxcrypt                  4.4.36        hd590300_1            98 KiB      conda  https://conda.anaconda.org/conda-forge/
libxkbcommon               1.13.1        hca5e8e5_0            818.3 KiB   conda  https://conda.anaconda.org/conda-forge/
libxml2                    2.15.3        h49c6c72_0            45.7 KiB    conda  https://conda.anaconda.org/conda-forge/
libxml2-16                 2.15.3        hca6bf5a_0            546.7 KiB   conda  https://conda.anaconda.org/conda-forge/
libxslt                    1.1.43        h711ed8c_1            239.7 KiB   conda  https://conda.anaconda.org/conda-forge/
libzlib                    1.3.2         h25fd6f3_2            62.1 KiB    conda  https://conda.anaconda.org/conda-forge/
matplotlib                 3.10.9        py312h7900ff3_0       17.4 KiB    conda  https://conda.anaconda.org/conda-forge/
matplotlib-base            3.10.9        py312he3d6523_0       7.9 MiB     conda  https://conda.anaconda.org/conda-forge/
munkres                    1.1.4         pyhd8ed1ab_1          15.5 KiB    conda  https://conda.anaconda.org/conda-forge/
ncurses                    6.6           hdb14827_0            897.4 KiB   conda  https://conda.anaconda.org/conda-forge/
numpy                      2.4.3         py312h33ff503_0       8.4 MiB     conda  https://conda.anaconda.org/conda-forge/
openjpeg                   2.5.4         h55fea9a_0            347.1 KiB   conda  https://conda.anaconda.org/conda-forge/
openldap                   2.6.13        hbde042b_0            767.7 KiB   conda  https://conda.anaconda.org/conda-forge/
openssl                    3.6.2         h35e630c_0            3 MiB       conda  https://conda.anaconda.org/conda-forge/
packaging                  26.2          pyhc364b38_0          89.4 KiB    conda  https://conda.anaconda.org/conda-forge/
pcre2                      10.47         haa7fec5_0            1.2 MiB     conda  https://conda.anaconda.org/conda-forge/
pillow                     12.2.0        py312h50c33e8_0       1015.2 KiB  conda  https://conda.anaconda.org/conda-forge/
pixman                     0.46.4        h54a6638_1            440.4 KiB   conda  https://conda.anaconda.org/conda-forge/
pthread-stubs              0.4           hb9d3cd8_1002         8.1 KiB     conda  https://conda.anaconda.org/conda-forge/
pyparsing                  3.3.2         pyhcf101f3_0          108.3 KiB   conda  https://conda.anaconda.org/conda-forge/
pyside6                    6.11.0        py312h50ac2ff_2       12.7 MiB    conda  https://conda.anaconda.org/conda-forge/
python                     3.12.0        hab00c5b_0_cpython    30.6 MiB    conda  https://conda.anaconda.org/conda-forge/
python-dateutil            2.9.0.post0   pyhe01879c_2          227.8 KiB   conda  https://conda.anaconda.org/conda-forge/
python_abi                 3.12          8_cp312               6.8 KiB     conda  https://conda.anaconda.org/conda-forge/
qhull                      2020.2        h434a139_5            540 KiB     conda  https://conda.anaconda.org/conda-forge/
qt6-main                   6.11.0        pl5321h16c4a6b_4      57.2 MiB    conda  https://conda.anaconda.org/conda-forge/
readline                   8.3           h853b02a_0            337 KiB     conda  https://conda.anaconda.org/conda-forge/
six                        1.17.0        pyhe01879c_1          18 KiB      conda  https://conda.anaconda.org/conda-forge/
tk                         8.6.13        noxft_h366c992_103    3.1 MiB     conda  https://conda.anaconda.org/conda-forge/
tornado                    6.5.5         py312h4c3975b_0       839.5 KiB   conda  https://conda.anaconda.org/conda-forge/
tzdata                     2025c         hc9c84f9_1            116.3 KiB   conda  https://conda.anaconda.org/conda-forge/
unicodedata2               17.0.1        py312h4c3975b_0       401 KiB     conda  https://conda.anaconda.org/conda-forge/
wayland                    1.25.0        hd6090a7_0            326.3 KiB   conda  https://conda.anaconda.org/conda-forge/
xcb-util                   0.4.1         h4f16b4b_2            20.3 KiB    conda  https://conda.anaconda.org/conda-forge/
xcb-util-cursor            0.1.6         hb03c661_0            20.3 KiB    conda  https://conda.anaconda.org/conda-forge/
xcb-util-image             0.4.0         hb711507_2            24 KiB      conda  https://conda.anaconda.org/conda-forge/
xcb-util-keysyms           0.4.1         hb711507_0            14 KiB      conda  https://conda.anaconda.org/conda-forge/
xcb-util-renderutil        0.3.10        hb711507_0            16.6 KiB    conda  https://conda.anaconda.org/conda-forge/
xcb-util-wm                0.4.2         hb711507_0            50.5 KiB    conda  https://conda.anaconda.org/conda-forge/
xkeyboard-config           2.47          hb03c661_0            389.9 KiB   conda  https://conda.anaconda.org/conda-forge/
xorg-libice                1.1.2         hb9d3cd8_0            57.3 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libsm                 1.2.6         he73a12e_0            26.9 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libx11                1.8.13        he1eb515_0            820 KiB     conda  https://conda.anaconda.org/conda-forge/
xorg-libxau                1.0.12        hb03c661_1            15 KiB      conda  https://conda.anaconda.org/conda-forge/
xorg-libxcomposite         0.4.7         hb03c661_0            14.1 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libxcursor            1.2.3         hb9d3cd8_0            31.8 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libxdamage            1.1.6         hb9d3cd8_0            12.9 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libxdmcp              1.1.5         hb03c661_1            20.1 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libxext               1.3.7         hb03c661_0            49.1 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libxfixes             6.0.2         hb03c661_0            19.6 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libxi                 1.8.2         hb9d3cd8_0            46.1 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libxrandr             1.5.5         hb03c661_0            29.7 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libxrender            0.9.12        hb9d3cd8_0            32.2 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libxtst               1.2.5         hb9d3cd8_3            32 KiB      conda  https://conda.anaconda.org/conda-forge/
xorg-libxxf86vm            1.1.7         hb03c661_0            18.3 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-xorgproto             2025.1        hb03c661_0            556.7 KiB   conda  https://conda.anaconda.org/conda-forge/
xz                         5.8.3         ha02ee65_0            23.8 KiB    conda  https://conda.anaconda.org/conda-forge/
xz-gpl-tools               5.8.3         ha02ee65_0            33.4 KiB    conda  https://conda.anaconda.org/conda-forge/
xz-tools                   5.8.3         hb03c661_0            93.7 KiB    conda  https://conda.anaconda.org/conda-forge/
zlib-ng                    2.3.3         hceb46e0_1            119.7 KiB   conda  https://conda.anaconda.org/conda-forge/
zstd                       1.5.7         hb78ec9c_6            587.3 KiB   conda  https://conda.anaconda.org/conda-forge/

Exercise: Try installing the package mtrf from Conda-Forge. What does the error message say?

Solution

pixi add mtrf

Error:   × failed to solve requirements of environment 'default' for platform 'linux-
  │ 64'
  ├─▶   × failed to solve the environment
  │   
  ╰─▶ Cannot solve the request because of: No candidates were found for mtrf
      *.

Exercise: Install the mtrf package from --pypi.

Solution

pixi add --pypi mtrf

✔ Added mtrf >=2.1.2, <3
Added these as pypi-dependencies.

Exercise: Open the pixi.toml file in VSCode and observe how the Conda-Forge and PyPI dependencies are listed separately.

NOTE: The pixi.toml file also contains other (user specific) information like author and platforms.

Solution

Content of pixi.toml: 

[workspace]
authors = ["atleer <atle.rimehaug@gmail.com>"]
channels = ["conda-forge"]
name = "my_project"
platforms = ["linux-64"]
version = "0.1.0"

[tasks]

[dependencies]
python = "==3.12"
numpy = ">=2.4.3,<3"
matplotlib = ">=3.10.9,<4"

[pypi-dependencies]
mtrf = ">=2.1.2, <3"

Section 3: Sharing and Reproducing Environments

Background

Pixi keeps track of all the dependencies we specify in our pixi.toml file. At the same time, it is keeping track of all packages in the virtual environment (including the dependencies of our dependencies) in the pixi.lock file.

The lock file allows us to exactly reproduce the pixi environment in a different directory or even on a different machine by simply copying the pixi.lock file and running pixi install.

Exercises

In the following exercises, you’ll reproduce the current pixi environment from the pixi.lock file. Here are the commands you need to know:

Code	Description
`mkdir new_dir/`	Create the directory `new_dir/`
`cp file.txt new_dir/`	Copy `file.txt` to `new_dir/` (Linux/MacOS)
`copy file.txt new_dir/`	Copy `file.txt` to `new_dir/` (Windows)
`cd new_dir`	Change the directory to `new_dir`
`cd ..`	Change the directory to the parent of the current directory
`pixi install`	Install the environment from `pixi.lock`

Exercise: Create a new directory called reproduce/ and copy pixi.toml and pixi.lock to this location. You can either use your file explorer or the terminal commands mkdir and cp.

Solution

mkdir /tmp/reproduce
cp pixi.toml /tmp/reproduce
cp pixi.lock /tmp/reproduce

Exercise: Move to the new reproduce/ directory and run pixi install to reproduce the environment there.

Solution

cd /tmp/reproduce
pixi install

 WARN Using local manifest /tmp/my_project/pixi.toml rather than /home/atleeri/repositories/learning_platform/new-learning-platform/pixi.toml from environment variable `PIXI_PROJECT_MANIFEST`
✔ The default environment has been installed.

Exercise: Run pixi list and check the installed version of numpy and ensure it matches the version in the original environment

Solution

pixi list

 WARN Using local manifest /tmp/my_project/pixi.toml rather than /home/atleeri/repositories/learning_platform/new-learning-platform/pixi.toml from environment variable `PIXI_PROJECT_MANIFEST`
Package                    Version       Build                 Size        Kind   Source
_openmp_mutex              4.5           20_gnu                28.3 KiB    conda  https://conda.anaconda.org/conda-forge/
alsa-lib                   1.2.15.3      hb03c661_0            571 KiB     conda  https://conda.anaconda.org/conda-forge/
array_api_compat           1.14.0                              196.8 KiB   pypi   array_api_compat-1.14.0-py3-none-any.whl
brotli                     1.2.0         hed03a55_1            19.6 KiB    conda  https://conda.anaconda.org/conda-forge/
brotli-bin                 1.2.0         hb03c661_1            20.5 KiB    conda  https://conda.anaconda.org/conda-forge/
bzip2                      1.0.8         hda65f42_9            254.1 KiB   conda  https://conda.anaconda.org/conda-forge/
ca-certificates            2026.4.22     hbd8a1cb_0            128 KiB     conda  https://conda.anaconda.org/conda-forge/
cairo                      1.18.4        he90730b_1            966.3 KiB   conda  https://conda.anaconda.org/conda-forge/
contourpy                  1.3.3         py312h0a2e395_4       312.9 KiB   conda  https://conda.anaconda.org/conda-forge/
cycler                     0.12.1        pyhcf101f3_2          14.4 KiB    conda  https://conda.anaconda.org/conda-forge/
cyrus-sasl                 2.1.28        hac629b4_1            205.2 KiB   conda  https://conda.anaconda.org/conda-forge/
dbus                       1.16.2        h24cb091_1            437.2 KiB   conda  https://conda.anaconda.org/conda-forge/
double-conversion          3.4.0         hecca717_0            70.1 KiB    conda  https://conda.anaconda.org/conda-forge/
font-ttf-dejavu-sans-mono  2.37          hab24e00_0            388.1 KiB   conda  https://conda.anaconda.org/conda-forge/
font-ttf-inconsolata       3.000         h77eed37_0            94.3 KiB    conda  https://conda.anaconda.org/conda-forge/
font-ttf-source-code-pro   2.038         h77eed37_0            684.4 KiB   conda  https://conda.anaconda.org/conda-forge/
font-ttf-ubuntu            0.83          h77eed37_3            1.5 MiB     conda  https://conda.anaconda.org/conda-forge/
fontconfig                 2.17.1        h27c8c51_0            264.4 KiB   conda  https://conda.anaconda.org/conda-forge/
fonts-conda-ecosystem      1             0                     3.6 KiB     conda  https://conda.anaconda.org/conda-forge/
fonts-conda-forge          1             hc364b38_1            4 KiB       conda  https://conda.anaconda.org/conda-forge/
fonttools                  4.62.1        py312h8a5da7c_0       2.8 MiB     conda  https://conda.anaconda.org/conda-forge/
freetype                   2.14.3        ha770c72_0            169.8 KiB   conda  https://conda.anaconda.org/conda-forge/
graphite2                  1.3.14        hecca717_2            97.3 KiB    conda  https://conda.anaconda.org/conda-forge/
harfbuzz                   14.2.0        h6083320_0            2.2 MiB     conda  https://conda.anaconda.org/conda-forge/
icu                        78.3          h33c6efd_0            12.1 MiB    conda  https://conda.anaconda.org/conda-forge/
keyutils                   1.6.3         hb9d3cd8_0            130.9 KiB   conda  https://conda.anaconda.org/conda-forge/
kiwisolver                 1.5.0         py312h0a2e395_0       75.3 KiB    conda  https://conda.anaconda.org/conda-forge/
krb5                       1.22.2        ha1258a1_0            1.3 MiB     conda  https://conda.anaconda.org/conda-forge/
lcms2                      2.19          h0c24ade_0            244.7 KiB   conda  https://conda.anaconda.org/conda-forge/
ld_impl_linux-64           2.45.1        default_hbd61a6d_102  710.9 KiB   conda  https://conda.anaconda.org/conda-forge/
lerc                       4.1.0         hdb68285_0            255.4 KiB   conda  https://conda.anaconda.org/conda-forge/
libblas                    3.11.0        6_h4a7cf45_openblas   18.2 KiB    conda  https://conda.anaconda.org/conda-forge/
libbrotlicommon            1.2.0         hb03c661_1            78.1 KiB    conda  https://conda.anaconda.org/conda-forge/
libbrotlidec               1.2.0         hb03c661_1            33.8 KiB    conda  https://conda.anaconda.org/conda-forge/
libbrotlienc               1.2.0         hb03c661_1            291.4 KiB   conda  https://conda.anaconda.org/conda-forge/
libcblas                   3.11.0        6_h0358290_openblas   18.2 KiB    conda  https://conda.anaconda.org/conda-forge/
libclang13                 22.1.4        default_h746c552_0    12.2 MiB    conda  https://conda.anaconda.org/conda-forge/
libcups                    2.3.3         h7a8fb5f_6            4.3 MiB     conda  https://conda.anaconda.org/conda-forge/
libdeflate                 1.25          h17f619e_0            71.8 KiB    conda  https://conda.anaconda.org/conda-forge/
libdrm                     2.4.125       hb03c661_1            303.5 KiB   conda  https://conda.anaconda.org/conda-forge/
libedit                    3.1.20250104  pl5321h7949ede_0      131.5 KiB   conda  https://conda.anaconda.org/conda-forge/
libegl                     1.7.0         ha4b6fd6_2            43.8 KiB    conda  https://conda.anaconda.org/conda-forge/
libegl-devel               1.7.0         ha4b6fd6_2            29.7 KiB    conda  https://conda.anaconda.org/conda-forge/
libexpat                   2.7.5         hecca717_0            74.8 KiB    conda  https://conda.anaconda.org/conda-forge/
libffi                     3.5.2         h3435931_0            57.2 KiB    conda  https://conda.anaconda.org/conda-forge/
libfreetype                2.14.3        ha770c72_0            7.9 KiB     conda  https://conda.anaconda.org/conda-forge/
libfreetype6               2.14.3        h73754d4_0            375.6 KiB   conda  https://conda.anaconda.org/conda-forge/
libgcc                     15.2.0        he0feb66_18           1017.4 KiB  conda  https://conda.anaconda.org/conda-forge/
libgcc-ng                  15.2.0        h69a702a_18           26.9 KiB    conda  https://conda.anaconda.org/conda-forge/
libgfortran                15.2.0        h69a702a_18           26.9 KiB    conda  https://conda.anaconda.org/conda-forge/
libgfortran5               15.2.0        h68bc16d_18           2.4 MiB     conda  https://conda.anaconda.org/conda-forge/
libgl                      1.7.0         ha4b6fd6_2            131.6 KiB   conda  https://conda.anaconda.org/conda-forge/
libgl-devel                1.7.0         ha4b6fd6_2            111.2 KiB   conda  https://conda.anaconda.org/conda-forge/
libglib                    2.88.0        h6705ce6_0            4.6 MiB     conda  https://conda.anaconda.org/conda-forge/
libglvnd                   1.7.0         ha4b6fd6_2            129.4 KiB   conda  https://conda.anaconda.org/conda-forge/
libglx                     1.7.0         ha4b6fd6_2            73.7 KiB    conda  https://conda.anaconda.org/conda-forge/
libglx-devel               1.7.0         ha4b6fd6_2            25.8 KiB    conda  https://conda.anaconda.org/conda-forge/
libgomp                    15.2.0        he0feb66_18           589.1 KiB   conda  https://conda.anaconda.org/conda-forge/
libiconv                   1.18          h3b78370_2            771.7 KiB   conda  https://conda.anaconda.org/conda-forge/
libjpeg-turbo              3.1.4.1       hb03c661_0            619 KiB     conda  https://conda.anaconda.org/conda-forge/
liblapack                  3.11.0        6_h47877c9_openblas   18.2 KiB    conda  https://conda.anaconda.org/conda-forge/
libllvm22                  22.1.4        hf7376ad_0            42.2 MiB    conda  https://conda.anaconda.org/conda-forge/
liblzma                    5.8.3         hb03c661_0            110.8 KiB   conda  https://conda.anaconda.org/conda-forge/
liblzma-devel              5.8.3         hb03c661_0            479.9 KiB   conda  https://conda.anaconda.org/conda-forge/
libnsl                     2.0.1         hb9d3cd8_1            32.9 KiB    conda  https://conda.anaconda.org/conda-forge/
libntlm                    1.8           hb9d3cd8_0            32.6 KiB    conda  https://conda.anaconda.org/conda-forge/
libopenblas                0.3.32        pthreads_h94d23a6_0   5.7 MiB     conda  https://conda.anaconda.org/conda-forge/
libopengl                  1.7.0         ha4b6fd6_2            49.6 KiB    conda  https://conda.anaconda.org/conda-forge/
libpciaccess               0.18          hb9d3cd8_0            27.8 KiB    conda  https://conda.anaconda.org/conda-forge/
libpng                     1.6.58        h421ea60_0            310.3 KiB   conda  https://conda.anaconda.org/conda-forge/
libpq                      18.3          h9abb657_0            2.7 MiB     conda  https://conda.anaconda.org/conda-forge/
libsqlite                  3.53.0        hf4e2dac_0            936.4 KiB   conda  https://conda.anaconda.org/conda-forge/
libstdcxx                  15.2.0        h934c35e_18           5.6 MiB     conda  https://conda.anaconda.org/conda-forge/
libstdcxx-ng               15.2.0        hdf11a46_18           26.9 KiB    conda  https://conda.anaconda.org/conda-forge/
libtiff                    4.7.1         h9d88235_1            425.1 KiB   conda  https://conda.anaconda.org/conda-forge/
libuuid                    2.42          h5347b49_0            39.4 KiB    conda  https://conda.anaconda.org/conda-forge/
libvulkan-loader           1.4.341.0     h5279c79_0            195.1 KiB   conda  https://conda.anaconda.org/conda-forge/
libwebp-base               1.6.0         hd42ef1d_0            419 KiB     conda  https://conda.anaconda.org/conda-forge/
libxcb                     1.17.0        h8a09558_0            386.6 KiB   conda  https://conda.anaconda.org/conda-forge/
libxcrypt                  4.4.36        hd590300_1            98 KiB      conda  https://conda.anaconda.org/conda-forge/
libxkbcommon               1.13.1        hca5e8e5_0            818.3 KiB   conda  https://conda.anaconda.org/conda-forge/
libxml2                    2.15.3        h49c6c72_0            45.7 KiB    conda  https://conda.anaconda.org/conda-forge/
libxml2-16                 2.15.3        hca6bf5a_0            546.7 KiB   conda  https://conda.anaconda.org/conda-forge/
libxslt                    1.1.43        h711ed8c_1            239.7 KiB   conda  https://conda.anaconda.org/conda-forge/
libzlib                    1.3.2         h25fd6f3_2            62.1 KiB    conda  https://conda.anaconda.org/conda-forge/
matplotlib                 3.10.9        py312h7900ff3_0       17.4 KiB    conda  https://conda.anaconda.org/conda-forge/
matplotlib-base            3.10.9        py312he3d6523_0       7.9 MiB     conda  https://conda.anaconda.org/conda-forge/
mtrf                       2.1.2                               68.1 KiB    pypi   mtrf-2.1.2-py3-none-any.whl
munkres                    1.1.4         pyhd8ed1ab_1          15.5 KiB    conda  https://conda.anaconda.org/conda-forge/
ncurses                    6.6           hdb14827_0            897.4 KiB   conda  https://conda.anaconda.org/conda-forge/
numpy                      2.4.3         py312h33ff503_0       8.4 MiB     conda  https://conda.anaconda.org/conda-forge/
openjpeg                   2.5.4         h55fea9a_0            347.1 KiB   conda  https://conda.anaconda.org/conda-forge/
openldap                   2.6.13        hbde042b_0            767.7 KiB   conda  https://conda.anaconda.org/conda-forge/
openssl                    3.6.2         h35e630c_0            3 MiB       conda  https://conda.anaconda.org/conda-forge/
packaging                  26.2          pyhc364b38_0          89.4 KiB    conda  https://conda.anaconda.org/conda-forge/
pcre2                      10.47         haa7fec5_0            1.2 MiB     conda  https://conda.anaconda.org/conda-forge/
pillow                     12.2.0        py312h50c33e8_0       1015.2 KiB  conda  https://conda.anaconda.org/conda-forge/
pixman                     0.46.4        h54a6638_1            440.4 KiB   conda  https://conda.anaconda.org/conda-forge/
pthread-stubs              0.4           hb9d3cd8_1002         8.1 KiB     conda  https://conda.anaconda.org/conda-forge/
pyparsing                  3.3.2         pyhcf101f3_0          108.3 KiB   conda  https://conda.anaconda.org/conda-forge/
pyside6                    6.11.0        py312h50ac2ff_2       12.7 MiB    conda  https://conda.anaconda.org/conda-forge/
python                     3.12.0        hab00c5b_0_cpython    30.6 MiB    conda  https://conda.anaconda.org/conda-forge/
python-dateutil            2.9.0.post0   pyhe01879c_2          227.8 KiB   conda  https://conda.anaconda.org/conda-forge/
python_abi                 3.12          8_cp312               6.8 KiB     conda  https://conda.anaconda.org/conda-forge/
qhull                      2020.2        h434a139_5            540 KiB     conda  https://conda.anaconda.org/conda-forge/
qt6-main                   6.11.0        pl5321h16c4a6b_4      57.2 MiB    conda  https://conda.anaconda.org/conda-forge/
readline                   8.3           h853b02a_0            337 KiB     conda  https://conda.anaconda.org/conda-forge/
six                        1.17.0        pyhe01879c_1          18 KiB      conda  https://conda.anaconda.org/conda-forge/
tk                         8.6.13        noxft_h366c992_103    3.1 MiB     conda  https://conda.anaconda.org/conda-forge/
tornado                    6.5.5         py312h4c3975b_0       839.5 KiB   conda  https://conda.anaconda.org/conda-forge/
tzdata                     2025c         hc9c84f9_1            116.3 KiB   conda  https://conda.anaconda.org/conda-forge/
unicodedata2               17.0.1        py312h4c3975b_0       401 KiB     conda  https://conda.anaconda.org/conda-forge/
wayland                    1.25.0        hd6090a7_0            326.3 KiB   conda  https://conda.anaconda.org/conda-forge/
xcb-util                   0.4.1         h4f16b4b_2            20.3 KiB    conda  https://conda.anaconda.org/conda-forge/
xcb-util-cursor            0.1.6         hb03c661_0            20.3 KiB    conda  https://conda.anaconda.org/conda-forge/
xcb-util-image             0.4.0         hb711507_2            24 KiB      conda  https://conda.anaconda.org/conda-forge/
xcb-util-keysyms           0.4.1         hb711507_0            14 KiB      conda  https://conda.anaconda.org/conda-forge/
xcb-util-renderutil        0.3.10        hb711507_0            16.6 KiB    conda  https://conda.anaconda.org/conda-forge/
xcb-util-wm                0.4.2         hb711507_0            50.5 KiB    conda  https://conda.anaconda.org/conda-forge/
xkeyboard-config           2.47          hb03c661_0            389.9 KiB   conda  https://conda.anaconda.org/conda-forge/
xorg-libice                1.1.2         hb9d3cd8_0            57.3 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libsm                 1.2.6         he73a12e_0            26.9 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libx11                1.8.13        he1eb515_0            820 KiB     conda  https://conda.anaconda.org/conda-forge/
xorg-libxau                1.0.12        hb03c661_1            15 KiB      conda  https://conda.anaconda.org/conda-forge/
xorg-libxcomposite         0.4.7         hb03c661_0            14.1 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libxcursor            1.2.3         hb9d3cd8_0            31.8 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libxdamage            1.1.6         hb9d3cd8_0            12.9 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libxdmcp              1.1.5         hb03c661_1            20.1 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libxext               1.3.7         hb03c661_0            49.1 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libxfixes             6.0.2         hb03c661_0            19.6 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libxi                 1.8.2         hb9d3cd8_0            46.1 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libxrandr             1.5.5         hb03c661_0            29.7 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libxrender            0.9.12        hb9d3cd8_0            32.2 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-libxtst               1.2.5         hb9d3cd8_3            32 KiB      conda  https://conda.anaconda.org/conda-forge/
xorg-libxxf86vm            1.1.7         hb03c661_0            18.3 KiB    conda  https://conda.anaconda.org/conda-forge/
xorg-xorgproto             2025.1        hb03c661_0            556.7 KiB   conda  https://conda.anaconda.org/conda-forge/
xz                         5.8.3         ha02ee65_0            23.8 KiB    conda  https://conda.anaconda.org/conda-forge/
xz-gpl-tools               5.8.3         ha02ee65_0            33.4 KiB    conda  https://conda.anaconda.org/conda-forge/
xz-tools                   5.8.3         hb03c661_0            93.7 KiB    conda  https://conda.anaconda.org/conda-forge/
zlib-ng                    2.3.3         hceb46e0_1            119.7 KiB   conda  https://conda.anaconda.org/conda-forge/
zstd                       1.5.7         hb78ec9c_6            587.3 KiB   conda  https://conda.anaconda.org/conda-forge/

Section 4: Running Python Scripts with Pixi Environments

Background

So far, we used Pixi to install Python as well as some packages but we never ran any code. In this section we are going to run a very simple Python script that simply prints "Hello World!" and run it in the Pixi environment we created.

To do this we could use pixi shell to activate the environment (as we did before) and then run the environment. However, we can make things even simpler and use the pixi run command which combines activating the environment and running the script in a single step.

Exercises

Code	Description
`pixi shell`	Activate the virtual environment
`python analyse.py`	Run the Python script `analyze.py`
`pixi run python analyze.py`	Run the Python script `analyze.py` in the current Pixi environment

Exercise: Copy the code in the cell below and save it to a file called hello.py.

print("Hello World!")

Hello World!

Exercise: Use pixi shell to activate the virtual environment. Then, run python hello.py to run the Python script.

NOTE: If hello.py is not located in your current directory you must specify the full path, e.g. my_directory/hello.py.

Solution

pixi shell
python hello.py

Hello World!

Exercise: Now use pixi run to execute the Python script without having to activate the environment first.

Solution

pixi run python hello.py

Hello World!

Section 5: Demo: Managing Multiple Environments

One strength of Pixi is its ability to manage projects that require multiple environments. This can be useful for research steps with multiple computation steps (e.g. preprocessing, statistics, visualization) that have different, incompatible dependencies. Pixi has two different concepts that are relevant here: features and environments. A feature is simply a set of packages with a specific name and an environment is a collection of features.

For this demo, lets create a little mock example. We’ll use the packages cowsay and pyfiglet in two separate environments. While there is no actual dependency conflict between these packages, it illustrates how we can isolate different parts of our code with dedicated environments.

First, we add the packages cowsay and pyfiglet and use the -f flag to assign each to a specific feature. Here, the features are called cow and fig. If you run the code you may get errors saying that these features do not belong to any environment - you can ignore these errors for now.

!pixi add --pypi -q -f cow cowsay
!pixi add --pypi -q -f fig pyfiglet

✔ Added cowsay
Added these as pypi-dependencies.
Added these only for feature: cow
✔ Added pyfiglet
Added these as pypi-dependencies.
Added these only for feature: fig

Let’s look at the content of pixi.toml. The dependencies listed in [dependencies] will be available for every environment while those listed under the [feature] headings will only be available in the environments that use these features.

!cat /tmp/my_project/pixi.toml

[workspace]
authors = ["atleer <atle.rimehaug@gmail.com>"]
channels = ["conda-forge"]
name = "my_project"
platforms = ["linux-64"]
version = "0.1.0"

[tasks]

[dependencies]
python = "==3.12"
numpy = ">=2.4.3,<3"
matplotlib = ">=3.10.9,<4"

[pypi-dependencies]
mtrf = ">=2.1.2, <3"

[feature.cow.pypi-dependencies]
cowsay = "*"

[feature.fig.pypi-dependencies]
pyfiglet = "*"

To create the environments we create a new heading called [environments] that define each environment as a list of features:

[environments]
env1=["cow"]
env2=["fig"]

Here, the environments are called env1 and env2 and contain the features "cow" and "fig" respectively. Each environment could also contain multiple features and the same feature can be used in multiple environments.

Now we can run cowsay in the environment env1 and pyfiglet in the environment env2. We specify the environment for the run command using the -e flag.

!pixi run -q -e env1 cowsay -t "Hello World"

  ___________
| Hello World |
  ===========
           \
            \
              ^__^
              (oo)\_______
              (__)\       )\/\
                  ||----w |
                  ||     ||

!pixi run -q -e env2 pyfiglet "Hello World"

 _   _      _ _        __        __         _     _ 
| | | | ___| | | ___   \ \      / /__  _ __| | __| |
| |_| |/ _ \ | |/ _ \   \ \ /\ / / _ \| '__| |/ _` |
|  _  |  __/ | | (_) |   \ V  V / (_) | |  | | (_| |
|_| |_|\___|_|_|\___/     \_/\_/ \___/|_|  |_|\__,_|