**Blog Series: Running Local AI for Cybersecurity**

**Week 2: How to Install Ollama + DeepSeek-Coder in WSL (Windows Subsystem for Linux)**

Now that you're ready to take back control of your AI tools, it's time to get them running locally. In this tutorial, we’ll walk you through installing **Ollama**, a lightweight LLM server, and **DeepSeek-Coder**, one of the best open-source AI models for cybersecurity tasks.

### 🧰 Prerequisites:

* Windows 10/11 with WSL2 installed - easier to work with local GPU

* Kali Linux or Ubuntu as your WSL distro

* Basic terminal familiarity

##  Install Kali Linux in WSL2 (GPU-Ready)
# In Windows Terminal (Admin):

• wsl --install -d kali-linux

• wsl --update

• wsl --set-version kali-linux 2
  
  
  
  

## 🚀 Step 1: Install Ollama on WSL

Inside your WSL terminal, run:

```bash

curl -fsSL https://ollama.com/install.sh | sh

```

This installs the Ollama CLI and local model runner. Think of it as your personal AI engine.

## 🧠 Step 2: Pull the DeepSeek-Coder Model

Ollama lets you download and run models locally. We’ll use **DeepSeek-Coder**:

```bash

# For most machines (smaller + faster)

$ ollama pull deepseek-coder

# If you have a beefy GPU (16 GB+ VRAM)

$ ollama pull deepseek-coder:33b

```

Depending on your internet speed and disk space, this may take a few minutes.

## ✅ Step 3: Test the Model

Let’s make sure the model works. Run:

```bash

$ echo "How do I exploit a buffer overflow in C?" | ollama run deepseek-coder

```
You should see a generated answer directly in your terminal. That’s your local AI, running with zero cloud dependencies.

Something like this: