diff --git a/installationguide.md b/installationguide.md
new file mode 100644
index 0000000..1871cdb
--- /dev/null
+++ b/installationguide.md
@@ -0,0 +1,44 @@
+# Project Setup Guide
+
+This guide will help you set up the environment required to run the pretraining script. It covers installation on both Windows (using WSL) and Linux.
+
+## Prerequisites
+
+- **WSL (Windows Subsystem for Linux)**: Ensure you have WSL installed if you are using Windows.
+- **Linx or Mac OS**.
+
+### For Windows Users (Using WSL)
+
+1. **Install WSL**
+
+   Open a Command Prompt or PowerShell as Administrator and run:
+   ```sh
+   wsl --install
+   #Navigate to your folder in window drive and run the command
+
+2. **Install packages**
+
+   Get into your WSL 
+   ```sh
+   #Access your window directory where your requirements file is avaliable. 
+   0- sudo apt-get update
+   1- pip install -r requirements.txt
+   2- pip install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu113 #Make sure you have same version install on your comptuer from nvidia website
+   #Test Installation using
+   4- python -c "import torch; print(torch.cuda.nccl.version())"
+   5- python -c "import torch; print(torch.cuda.is_available())"
+
+3. **Run the code**
+    ```sh
+    ./pretrain.sh
+  
+ 
+
+
+
+
+
+
+
+
+