diff --git a/03_instructional_team/markdown_slides/images/colab_dropdown.png b/03_instructional_team/markdown_slides/images/colab_dropdown.png
new file mode 100644
index 000000000..bd835b094
Binary files /dev/null and b/03_instructional_team/markdown_slides/images/colab_dropdown.png differ
diff --git a/03_instructional_team/markdown_slides/images/colab_setup.png b/03_instructional_team/markdown_slides/images/colab_setup.png
new file mode 100644
index 000000000..a7dfb28e4
Binary files /dev/null and b/03_instructional_team/markdown_slides/images/colab_setup.png differ
diff --git a/03_instructional_team/markdown_slides/images/vs_code_ntbk.png b/03_instructional_team/markdown_slides/images/vs_code_ntbk.png
new file mode 100644
index 000000000..d3860dd5b
Binary files /dev/null and b/03_instructional_team/markdown_slides/images/vs_code_ntbk.png differ
diff --git a/SETUP.md b/SETUP.md
index c6cc61f1c..9d8287ff2 100644
--- a/SETUP.md
+++ b/SETUP.md
@@ -1,5 +1,7 @@
 # Setup
-## Environment Setup Guide
+
+
+## Virtual Environment Setup Guide
 Before using this repo, make sure you’ve completed the [environment setup guide](https://github.com/UofT-DSI/onboarding/blob/main/environment_setup/README.md), which installs the core tools you’ll need for this module, such as:
 
 - Git  
@@ -7,7 +9,7 @@ Before using this repo, make sure you’ve completed the [environment setup guid
 - Visual Studio Code
 - UV
 
-## Necessary Packages
+### Necessary Packages
 The Deep Learning module uses its own isolated environment called `deep-learning-env` so that packages don’t conflict with other projects. 
 We use UV to create this environment, activate it, and install the required packages listed in the module’s `pyproject.toml`.  
 This setup only needs to be done **once per module**, after that, you just activate the environment whenever you want to work in this repo.  
@@ -35,7 +37,7 @@ Open a terminal (macOS/Linux) or Git Bash (Windows) in this repo, and run the fo
     uv sync --active
     ```
 
-## Environment Usage
+### Environment Usage
 In order to run any code in this repo, you must first activate its environment.
 - for macOS/Linux:
     ```
@@ -60,7 +62,33 @@ deactivate
 
 > **👉 Remember**   
 > Only one environment can be active at a time. If you switch to a different repo, first deactivate this one (or just close the terminal) and then activate the new repo’s environment.
+> 
+### Environments and Notebooks
+
+![kernel selection](03_instructional_team/markdown_slides/images/vs_code_ntbk.png)
+
+To use your virtual environment in a notebook, click the kernel selection dropdown, select `Python Environments`, then click the `deep-learning-env` from the dropdown. Then restart your notebook.
+
+
+## Colab
+An alternative to setting up your virtual environment is to instead use Google Colab. Colab is a free, cloud-based Jupyter notebook environment provided by Google that runs entirely in your browser. It requires minimal, just a Google account. Colab provides access to GPUs and TPUs, which can significantly speed up training deep learning models. 
+
+You can access a Colab kernel for your notebooks by installing the Colab extension in VSCode:
+
+![colab setup guide](03_instructional_team/markdown_slides/images/colab_setup.png)
+
+Once you have installed this kernel, you can click the kernel dropdown in your notebook, click `Select Another Kernel`, and select `Colab`. You will see a dropdown 
+
+![colab dropdown](03_instructional_team/markdown_slides/images/colab_dropdown.png)
+
+This will ask you to log into a Google account, simply follow the dropdown. For normal usage, click `Auto-Connect`. Otherwise click `New Colab Server` and select `CPU`,`GPU`, or `TPU`. For this course, using `GPU` may be useful if you have not installed GPU drivers. 
+
+**Pros and Cons** Colab is very convenient as it has many packages pre-installed. However, there are some downsides: 
+* This does not work offline and requires you to be connected to the internet. 
+* Colab is limited to notebooks and you cannot utilize it to run scripts. 
+* This is run on Google's servers, making it unsuitable for applications where data security is important. 
 
 ---
 
-For questions or issues, please contact the Deep Learning Module learning support team or email courses.dsi@utoronto.ca.
\ No newline at end of file
+
+For questions or issues, please contact the Deep Learning Module learning support team or email courses.dsi@utoronto.ca.