Back to Blog

Difference between load, require, and require_relative in Ruby

Aug 31, 2026
•3 min read

Ruby provides a few different ways to load code from other files. In this article, we are going to explore three methods: load, require, and require_relative and understand when to use each of them.

This article is Post #6 of the 100 Days to Offload challenge.

Now, before moving forward, we need to know about these two global variables provided by the Ruby process.

  1. $LOAD_PATH
  • It is a global variable that contains the list of directories Ruby searches when using require.
  1. $LOADED_FEATURES
  • It is a global variable that contains the files or libraries/gems that Ruby has already loaded using require.
Note

We can look for global variables in irb using global_variables, and in Pry we can also use ls -g.

Then we can check both global variables that are arrays using the puts command. puts $LOAD_PATH and puts $LOADED_FEATURES.

require

require is used when we want to load files once during the lifetime of a Ruby process.

When we use require, the following things can happen conceptually:

  • Ruby checks in the $LOADED_FEATURES array if the file or library has already been loaded or not.
    • If loaded, do nothing.
    • Else Ruby searches the required file in the $LOAD_PATH array.
      • If found, then the file is loaded & added to the $LOADED_FEATURES array.
      • If not found, then Ruby raises a LoadError.

Uses:

We use require to load library or gem. For example:

require "json"

or

require "rails"
Note

We may also use require to load our own project files, but those files should be in the $LOAD_PATH array. It creates a dependency on $LOAD_PATH in order to use require to load our own project files.

To load files from our own project, we use require_relative.

require_relative

require_relative is similar to require, but there is a subtle difference. This method is useful to require files relative to the file containing the require_relative call without having to change $LOAD_PATH.

Uses:

  • Use require_relative to load files from our own project.

For example, we have a file structure like this:

project/
  ├── app/
  │   └── hello.rb
  └── main.rb

Then in main.rb we can use require_relative to load hello.rb like this:

require_relative "app/hello"
Note

When building a gem, we conventionally put Ruby code in the lib directory. So when the gem is used, the lib directory is added to $LOAD_PATH and we can use require to load files from the gem.

Load

The load method is the least used method to load a file. It is useful in irb/pry to reload files we are working on.

This method loads a file every time we call it.

Unlike require and require_relative, the load method does not check $LOADED_FEATURES to see if the file is already loaded.

That's why require loads the file only once, but the load method loads files every time we call it.

Another difference between load and require is with load we provide the complete file name with extension like .rb, but with require, the file extension is not required.

for example

load "app/hello.rb"
 
 
require "app/hello"

This was a small concept but worth learning. This will be helpful to know when working with gems or projects in Ruby.

The main thing I'll remember is:

  • Use require and require_relative to load files once during the lifetime of a Ruby process.
  • Use require_relative to load files relative to the current file.
  • load is useful to reload files we are working on in irb/pry.