Prerequisites
=============

Before you begin, make sure you have the following requirements installed and configured:

1. A MacOS laptop (primary support)
2. Lambda Labs API key
3. Hugging Face access token

**OS Support:**

- **MacOS**: Full support with automatic dependency installation
- **Other OS (Windows/Linux)**: Supported if you have `docker compose` installed manually

.. note::
  For non-MacOS users, you'll need to install `docker compose` manually before proceeding with the installation.


1. Get a Lambda Labs API Key
----------------------------

You will need a Lambda Labs API key to launch and manage cloud GPU instances.

To generate your API key:

#. Go to https://lambda.ai/ and click **Sign Up** (or **Log In** if you already have an account).
#. After signing in, navigate to the **API Keys** tab in the left sidebar.
#. Click **Generate API Key** and give it a name.
#. Your key will look something like:

   .. code-block:: text

      secret_xxx

#. Be sure to save your key in a safe place. You will not be able to retrieve it again, and you will need to generate a new one if you lose access.

2. Get a Hugging Face Access Token
----------------------------------

You need a Hugging Face token to access models from the Hugging Face Hub.

To get your HF access token:

#. Go to https://huggingface.co/login and **Sign Up** (or **Log In** if you already have an account).
#. Then go to: https://huggingface.co/settings/tokens.
#. Click **+ Create new token**, choose the token type **Read**, give it a name (e.g., "lmignite"), and click **Create token**.
#. Your token will look something like:

   .. code-block:: text

      hf_xxx

#. Be sure to save your token somewhere safe. You will not be able to see it again after closing the modal dialog. If you lose it, you’ll need to create a new one.

.. note::
   Some models on Hugging Face are "gated" and require you to request access on the model card page. You may need to agree to conditions, and approval can be automatic or manual depending on the model author.

Done!
-----

Once you've completed these three steps, you're ready to proceed with the rest of the tutorial.