Skip to content

Install Hummingbot from Source

Re-compiling files

If you make changes to the code, make sure to re-compile the code with ./compile to ensure that any changes to Cython files are compiled before running Hummingbot

You can install Docker and Hummingbot by selecting the following options below:

  • Scripts: download and use automated install scripts
  • Manual: run install commands manually


Supported versions: 16.04 LTS, 18.04 LTS, 19.04

    # 1) Download install script

    # 2) Enable script permissions
    chmod a+x

    # 3) Run installation
# 1) Install dependencies
sudo apt-get update
sudo apt-get install -y build-essential

# 2) Install Miniconda3

# 3) Reload .bashrc to register "conda" command
exec bash

# 4) Clone Hummingbot
git clone

# 5) Install Hummingbot
cd hummingbot && ./clean && ./install

# 6) Activate environment and compile code
conda activate hummingbot && ./compile

# 7) Start Hummingbot


Xcode command line tools

Running Hummingbot on macOS requires Xcode and Xcode command line tools.

xcode-select --install


Hummingbot requires Python 3 and other Python libraries. To manage these dependencies, Hummingbot uses Anaconda, an open-source environment, and package manager for Python that is the current industry standard for data scientists and data engineers.

To install Anaconda, go to the Anaconda site and download the Python 3.7 installer for your operating system. Both the graphical installer and the command line installer will work. Run the installer, and it will guide you through the installation process.

Afterward, open a Terminal window and try the conda command. If the command is valid, then Anaconda has been successfully installed, even if the graphical installer says that it failed.


If you use ZSH or another Unix shell, copy the code snippet below to your .zshrc or similar file. By default, Anaconda only adds it to your .bash_profile file. This makes the conda command available in your root path.

__conda_setup="$(CONDA_REPORT_ERRORS=false '/anaconda3/bin/conda' shell.bash hook 2> /dev/null)"
if [ $? -eq 0 ]; then
    \eval "$__conda_setup"
    if [ -f "/anaconda3/etc/profile.d/" ]; then
        . "/anaconda3/etc/profile.d/"
        CONDA_CHANGEPS1=false conda activate base
        \export PATH="/anaconda3/bin:$PATH"
unset __conda_setup


# 1) Clone Hummingbot repo
git clone

# 2) Navigate into the hummingbot folder
cd hummingbot

# 3) Run install script

# 4) Activate the environment
conda activate hummingbot

# 5) Compile

# 6) Run Hummingbot



  1. Install Git for Windows.

  2. During the installation of Git Bash, make sure to tick the Enable experimental support for pseudo consoles.


  1. Install Python for Windows.
  2. Install Anaconda or miniconda.
  3. Install Visual Studio Code, Visual Studio BuildTools 2019, Core Features, C++ and C++ redistributable 2019.

  4. During installation, make sure to add Anaconda or Miniconda to your PATH environmental variable by clicking the tick box as shown below, or you can add them manually. anaconda-path


Some prerequisites are large applications and may need to restart your computer.


  1. Launch Git Bash App.
  2. Run the following commands:
# initialized conda
conda init bash
# exit git-bash to take effect
# launch Git Bash App again
  1. You can install Hummingbot by selecting either of the following options from the tabs below:

  2. Scripts: download and use automated install scripts.

  3. Manual: run install commands manually.
# 1) Navigate to the root folder
cd ~

# 2) Download install script
curl -o

# 3) Enable script permissions
chmod a+x

# 4) Run installation
cd ~
export CONDAPATH="$(pwd)/miniconda3"
export PYTHON="$(pwd)/miniconda3/envs/hummingbot/python3"
# Clone Hummingbot
git clone
# Install Hummingbot
export hummingbotPath="$(pwd)/hummingbot" && cd $hummingbotPath && ./install
# Activate environment and compile code
conda activate hummingbot && ./compile
# Start Hummingbot
winpty python bin/

Last update: 2021-12-22
Back to top