1 Getting Started with Claude Code + Ollama (Windows)

Because copying commands is faster than explaining why node_modules is 2 GB.

Step 1: Install Claude Code CLI

Choose the command that matches your terminal.

Windows PowerShell

irm https://claude.ai/install.ps1 | iex

Windows Command Prompt (CMD)

curl -fsSL https://claude.ai/install.cmd -o install.cmd && install.cmd && del install.cmd

Official Documentation

Claude Code Quickstart: https://code.claude.com/docs/en/quickstart

Getting Started: https://support.claude.com/en/articles/14552382-your-first-day-in-claude-code

Step 2: Install Ollama

Open PowerShell and run:

irm https://ollama.com/install.ps1 | iex

Official Documentation

https://docs.ollama.com/quickstart

https://docs.ollama.com/windows

Step 3: Create a Project Folder

Navigate to the drive where you'd like to keep your projects.

cd D:\

Using Command Prompt (CMD)

mkdir MyFolder

Using PowerShell

New-Item -ItemType Directory -Path "MyFolder"

Move into the new folder:

cd MyFolder

Pro Tip: Name the folder something meaningful. Six months from now, New Folder (17) won't spark any memories.

Step 4: Log in to Ollama

Before launching a cloud model, sign in to your Ollama account.

ollama signin

Your browser will open automatically to complete authentication. Once you're signed in, return to your terminal and continue with the next step.

Why this step?

Cloud-hosted models require an authenticated Ollama account. The good news? It's a one-time setup—future you will appreciate present you.

Step 5: Launch Claude Code with an Ollama Cloud Model

Run the following command:

ollama launch claude --model qwen3-coder-next:cloud

That's it! Claude Code will launch using the Qwen3 Coder Next (Cloud) model through Ollama.

For more launch options and supported integrations, see the official Ollama documentation:

https://docs.ollama.com

✅ Quick Checklist

✅ Claude Code CLI installed
✅ Ollama installed
✅ Project directory created
✅ Logged in to Ollama
✅ Terminal opened inside your project
✅ Claude Code launched with the desired model

Happy coding! If something breaks, don't panic—it's probably just the terminal expressing its feelings. 😄