mamba

Create a new environment, installing the specified packages into it

$ mamba create [[-n|--name]] [environment_name] [python=3.10 matplotlib]

$ mamba install [[-c|--channel]] [conda-forge] [python=3.6 numpy]

$ mamba update --all

$ mamba repoquery search [numpy]

$ mamba info --envs

$ mamba clean -pt

$ mamba activate [environment_name]

$ mamba list

mamba [global options] [command options]

install
    Installs packages into an environment.

update
    Updates packages in an environment.

remove
    Removes packages from an environment.

create
    Creates a new environment.

list
    Lists packages in an environment.

search
    Searches for packages in available channels.

env create
    Creates an environment from an environment.yml file.

-n, --name ENV_NAME
    Name of the environment to use.

-y, --yes
    Do not ask for confirmation.

-c, --channel CHANNEL
    Channel to search for packages. Can be specified multiple times.

--prefix PATH
    Full path to environment location (instead of using 'name').

--offline
    Don't connect to the internet.

Mamba is a fast, robust, and cross-platform package manager. It is a reimplementation of the conda package manager in C++. Mamba aims to provide a significantly faster experience, especially when solving complex dependency graphs. It is fully compatible with conda packages, environments, and most of its commands. This allows users to seamlessly transition to using mamba for everyday package management tasks like installing, updating, and removing packages. The main benefit of mamba is its parallel package downloads and constraint solving, which considerably reduces the time required to manage environments, especially for large projects with numerous dependencies.

Mamba provides a command-line interface (CLI) similar to conda, making it easy for existing conda users to adapt. It also offers improved error reporting and enhanced user experience, contributing to a more efficient workflow for managing software packages and dependencies.

While largely compatible, certain less common conda features might not be fully implemented in mamba. Always test critical workflows to ensure desired behavior.

Mamba uses channels to locate packages. Common channels include 'defaults', 'conda-forge', and others. Specifying the correct channels is crucial for finding the desired packages.

Mamba environments allow you to isolate project dependencies. Creating separate environments prevents conflicts between different projects using different versions of the same packages.

Mamba was created to address performance bottlenecks in conda's dependency solving process. Its development focused on leveraging C++ for faster execution and parallelization. The project has gained significant traction in the scientific computing and data science communities due to its speed and reliability, becoming a preferred alternative to conda for many users. The development is lead by QuantStack and has many contributors from the conda-forge community.

conda(1), pip(1)