conda-activate

Activate a Conda environment

TLDR

Activate an existing environment named myenv

$ conda activate myenv

Activate an existing environment located at custom path

$ conda activate [path/to/myenv]

Stack myenv environment on top of a previous environment making libraries/commands/variables from both accessible

$ conda activate --stack myenv

Start a clean environment myenv without stacking it making previous environment libraries/commands/variables not accessible

$ conda activate --no-stack myenv

Display help

$ conda activate [[-h|--help]]

SYNOPSIS

conda activate [-h] [--dry-run] [--no-hoist] [--stack] [--no-stack] [--reverse] [prefix]

-h, --help
    Show help message and exit

--dry-run
    Display commands without executing changes to PATH or PS1

--no-hoist
    Disable package hoisting for faster activation (ignores build shortcuts)

--stack
    Push current environment onto stack before activating new one

--no-stack
    Prevent stacking; deactivate current before activating new

--reverse
    Reverse last stacking operation (deactivate top environment)

prefix
    Environment name or full path to prefix (defaults to base if omitted)

DESCRIPTION

The conda activate command (often invoked via shell hooks as part of Conda's initialization) activates a specified Conda environment, modifying the current shell session to use packages and binaries from that environment. It prepends the environment's bin directory to the PATH, updates shell prompt (PS1) to indicate the active environment, and sets environment variables like CONDA_DEFAULT_ENV, CONDA_PREFIX, and CONDA_PROMPT_MODIFIER.

This enables isolated Python interpreters, libraries, and tools without system-wide installation. Unlike older source activate, conda activate works reliably across shells (bash, zsh, fish, etc.) via init scripts in $CONDA_PREFIX/etc/profile.d/conda.sh. Activation is non-destructive; conda deactivate restores the previous state. Stacking allows multiple environments, useful for dependency hell scenarios.

Requires Conda shell integration: run conda init once. Note: 'conda-activate' may refer to internal hook scripts like those in etc/profile.d, but user-facing is conda activate. Supports base environment activation too.

CAVEATS

Requires shell initialization via conda init <shell>; fails silently in non-interactive shells. Environment names are case-sensitive. Stacking limited by shell recursion depth. Not for Windows cmd.exe (use conda-activate.bat).

SHELL HOOKS

Activation sources $CONDA_PREFIX/etc/profile.d/conda.sh, defining __conda_activate function. Manual sourcing enables in custom shells.

STACK MANAGEMENT

Use conda activate --stack for layered envs; view stack with conda info --envs (active marked with *).

HISTORY

Introduced in Conda 4.6 (2018) as robust replacement for source activate/deactivate scripts, which were unreliable across shells. Developed by Anaconda Inc. to support multi-shell integration via hooks. Enhanced in 4.8+ with stacking and hoisting optimizations. Now standard in Miniconda/Anaconda distributions.