Getting Started


These instructions are tempporarily for a release-cantidate. They will be changed on January 25th with the launch of Phase 3.

This section will walk you through the requirements needed to run a Coda protocol node on your local machine and connect to the network.


This documentation is for the beta release. The commands and APIs may change before the initial release. Last updated for version 0.0.10-beta.


Software: macOS or Linux (currently supports Debian 9 and Ubuntu 18.04 LTS)

Note: Windows is not officially supported at this time. However, community members succeeded in setting up nodes using Windows Subsystem for Linux. Click here for instructions on using Windows, which are created by the community.

Hardware: Sending and receiving coda does not require any special hardware, but participating as a node operator currently requires:

  • at least a 4-core processor
  • at least 16 GB of RAM

GPUs aren't currently required, but may be required for node operators when the protoctol is upgraded.

Network: At least 1 Mbps connection

VM Instances: O(1) Labs has tested running nodes on several cloud providers, and recommends the following instances for basic node operator needs. Keep in mind that custom requirements as well as different cost constraints may require a different instance type.


The newest binary releases can be found below. Instructions are provided for macOS and Linux below:

This is a large download, around 1GB, so the install might take some time.


If you installed coda from a previous release, you'll need to upgrade it so that you won't get banned by the network for using an older client. See instructions below for upgrading both macOS and Linux builds.


Install using Homebrew.

brew install codaprotocol/coda/coda

If you already have coda installed from a previous release, run:

brew upgrade coda

You can run coda -help to check if the installation succeeded.

Ubuntu 18.04 / Debian 9

For Linux distrobutions we are temporarily using Docker. Run the following command to install Docker.

curl get.docker.com | sh

Next to download, start and connect to the image by running the command below:

docker run -it --entrypoint /bin/bash  codaprotocol/coda-daemon:0.0.11-beta1-release-0.0.12-beta-493b4c6

All the following commands should be run inside the Docker container.

You can run coda -help to check if the installation succeeded.


Windows is not yet supported. If you have any interest in developing Coda for Windows, please reach out to support@o1labs.org or reach out in the Discord server.

Build from source

If you're running another Linux distro or a different version of macOS, you can try building Coda from source code. Please note that other operating systems haven't been tested thoroughly, and may have issues. Feel free to share any logs and get troubleshooting help in the Discord channel.

Set up port forwarding and any firewalls

If you are running a firewall, you should allow traffic on TCP port 8303. Additionally, unless the -external-ip YOUR_IP flag is provided, the daemon will use HTTPS (443) and HTTP (80) to try and determine its own IP address.

You may need to configure your router's port forwarding to allow inbound traffic to the following ports through your external IP address.

  • TCP port 8302 and 8303

For walk-through instructions see this guide.


Now that you've installed Coda and configured your network, let's move on to the fun part - sending a transaction!