Skip to main content

Before you begin

Using Anaconda Platform (Cloud) requires membership to an organization with a current subscription. If you’re working with an Anaconda representative, they will create and subscribe your organization for you. To get started:
  1. Install conda on your computer.
  2. Authenticate with anaconda login.
  3. Configure access to your organization’s .
Once conda is configured, you can start building for your projects using secure, approved .

Installing conda

You can obtain conda by installing either Anaconda Distribution or Miniconda.
If you already have conda installed, skip ahead to Authenticating to Anaconda Platform.
If you do not have conda installed, install it now and return to this page when you’re done.

Authenticating to Anaconda Platform

You must be assigned a seat within your organization before you can authenticate. If you have not been assigned a seat, contact your organization administrator.
  1. Open Anaconda Prompt (Terminal on macOS/Linux).
  2. Verify that anaconda-auth is installed:
    conda list --name base anaconda-auth
    
    If the output does not display package information, install it:
    conda install --name base anaconda-auth
    
  3. Log in to Anaconda:
    anaconda login
    
    A browser window opens to complete the authentication process. If your organization has configured SSO, you are redirected to your company’s SSO login page.
    If you belong to more than one organization, use the arrow keys to select the organization you want to authenticate to.

API key storage

After authentication completes, an API key is generated and stored in your keyring:
C:\Users\<USERNAME>\.anaconda\keyring

Environment logging setup

If your organization has environments enabled, anaconda login prompts you to install the necessary plugins for environment logging and scanning. After setup, conda automatically logs any environment you create or update to your registered organization.
(base) ➜  ~ anaconda login
Anaconda Environment Manager is required by your organization. It is recommended to install. Proceed? [y/n] (y): y
Installing anaconda-env-manager...
✔︎ anaconda-env-manager installed successfully.
Only one organization found, automatically selecting: my-organization
Linking this machine to organization my-organization...
✔ This machine is now registered with my-organization. Conda environments will be logged to this organization.
If you belong to multiple organizations, you are prompted to select one organization from the list before registration completes.
For more information, see Environment logging and scanning.

Issuing and setting your token

Your organization access token grants conda access to your organization’s channels and packages. To issue and set your token, run the following command:
anaconda token install
If you belong to more than one organization, you are prompted to select which organization to issue a token for.

Configuring conda to use organization channels

To access packages from your organization’s channels, add them to your .condarc file:
  1. Log in to anaconda.com/app.
  2. Select Channels from the left-hand navigation.
  3. Select Copy channel path.
    Copy channel path button
  4. Open Anaconda Prompt (Terminal on macOS/Linux) and run the following command:
    conda config --prepend channels <CHANNEL_PATH>
    
    This adds the specified channel to the top of your channels: list, giving it top priority when conda searches for packages.
    Example .condarc file
    channel_settings:
      - channel: https://repo.anaconda.cloud/*
        auth: anaconda-auth
    channels:
      - https://repo.anaconda.cloud/repo/<ORG_ID>/<CHANNEL_NAME>
      - defaults
    add_anaconda_token: true
    default_channels:
      - https://repo.anaconda.cloud/repo/main
    
    If you want to use your organization’s channels exclusively, ensure they are the only channels in your .condarc’s channels: list.
For more information about managing channels, see Channels.

Using Anaconda behind a firewall or proxy (Optional)

Some companies have security policies that prevent communications on their network with external servers like Anaconda. Under these circumstances, you’ll need to connect to your company’s firewall/proxy server in order to download packages successfully. To connect to a firewall/proxy server, you’ll need to include a proxy_servers: section in the .condarc file that contains the URL to the proxy server. This entry must also contain a username and password for logging in to the proxy server. Speak with your IT Administrator if you do not have this information. There are no commands to include this portion of the .condarc file, so you need to manually include the following lines:
If your password contains special characters, you’ll need to escape them using percent encoding as described here.
proxy_servers:
  http: http://<USERNAME>:<PASSWORD>@<URL>:8080
  https: https://<USERNAME>:<PASSWORD>@<URL>:8443
You’ll also need to work with your IT team to allowlist connections to the main package repositories once you’ve configured your connection to the firewall/proxy server. The main package repositories are:
https://conda.anaconda.org
https://repo.anaconda.com
https://repo.anaconda.cloud
In some situations, it is necessary to export the HTTP_PROXY and HTTPS_PROXY environment variables to utilize the proxy server. To export your environment variables, open a terminal and run the following commands:
export HTTP_PROXY=http://<USERNAME>:<PASSWORD>@<URL>:8080
export HTTPS_PROXY=https://<USERNAME>:<PASSWORD>@<URL>:8443
For more information about using conda behind a proxy server, see Configure conda for use behind a proxy server.

Verifying your configuration

Test your configuration by creating an environment and installing a package:
  1. Create a test environment:
    conda create --name <ENV_NAME>
    
  2. Verify the environment was created:
    conda env list
    
  3. Activate the environment:
    conda activate <ENV_NAME>
    
  4. Install a package:
    conda install <PKG_NAME>
    
  5. If necessary, delete the environment to clean up:
    conda env remove --name <ENV_NAME>