As a student, you need good tools without paying for subscriptions. In this tutorial, I'll show step by step how to:
- get JetBrains IDEs for free (IntelliJ IDEA, WebStorm...) with the Student Pack;
- install Ollama to run an AI model locally;
- connect Claude Code to that local model, so you get a coding assistant with no API costs.
Everything here is done on Windows 11 / PowerShell. The steps are almost identical on macOS and Linux.
Prerequisites
- An active university email (mine is
@usmba.ac.ma) or another proof of student status. - A good internet connection: the IDEs are ~1 GB each and the model is ~3 GB.
- 16 GB of RAM (what I used for this tutorial). 8 GB should be enough for a small 4B model like Qwen 3.5 4B, but it may be slower.
Part 1: Free JetBrains with the Student Pack
1.1 Apply for the student license
- Go to the student pack page: https://www.jetbrains.com/academy/student-pack/
- Start the license request and create (or sign in to) your JetBrains account.
- Verify your student status. The easiest way is with your university email address. You'll receive a confirmation email: click the link inside.
According to JetBrains' terms, the student license is for non-commercial use (studies, personal projects, learning) and is renewable as long as you remain a student. Always check the current terms on the official site.
Once activated, the license appears in your JetBrains account and gives you access to all the IDEs.
1.2 Download the IDEs
I'll use two examples: WebStorm (JavaScript / web) and IntelliJ IDEA (Java / Kotlin).
WebStorm: https://www.jetbrains.com/webstorm/download/
IntelliJ IDEA: https://www.jetbrains.com/idea/download/
Choose your system (Windows, macOS or Linux) and click Download. On Windows you get an .exe file.
1.3 Install
Run the downloaded .exe files.
Follow the wizard (Next, Next, Install). Installation takes a few minutes.
1.4 Activate the license
On first launch, sign in with your JetBrains account (the one with the student license) and the IDE activates automatically. That's it: professional IDEs, for free.
Part 2: Install Ollama
Ollama lets you download and run language models on your own machine. Your code never leaves your computer and there is no per-request cost.
2.1 (Optional) Create an account
On https://ollama.com you can create a free account. The dashboard shows your free quota for cloud models. For the rest of this tutorial (100% local), it isn't required.
2.2 Install Ollama
Go to the download page: https://ollama.com/download
On Windows, open PowerShell and run:
irm https://ollama.com/install.ps1 | iex
Wait for the installation to finish:
You can also click "Download manually" and use the graphical installer.
2.3 First launch
Simply type:
ollama
A welcome screen appears. Press Enter to continue.
2.4 Check the installation
ollama --version
Tip: you can run this directly in the built-in terminal of IntelliJ or WebStorm.
2.5 Download a model
I chose Qwen 3.5 4B, a small open-source model (~3 GB) that runs on a regular laptop:
ollama pull qwen3.5:4b
When you see success, the model is ready. You can test it:
ollama run qwen3.5:4b
Part 3: Connect Claude Code to Ollama
Claude Code is a coding assistant that runs in your terminal. Ollama can talk to it thanks to an Anthropic-compatible API: Claude Code thinks it's talking to its usual server, but your local model is answering.
3.1 The easy way: ollama launch
In your project folder (for example in the IntelliJ terminal):
ollama launch claude --model qwen3.5:4b
Ollama installs Claude Code if it isn't there yet, sets the required variables for you and starts it with the chosen model.
And there it is: Claude Code running with qwen3.5:4b locally.
3.2 The manual way (to understand what's happening)
If Claude Code is already installed, you can point it to Ollama yourself with three environment variables.
PowerShell:
$env:ANTHROPIC_AUTH_TOKEN = "ollama"
$env:ANTHROPIC_API_KEY = ""
$env:ANTHROPIC_BASE_URL = "http://localhost:11434"
claude --model qwen3.5:4b
macOS / Linux:
export ANTHROPIC_AUTH_TOKEN=ollama
export ANTHROPIC_API_KEY=""
export ANTHROPIC_BASE_URL=http://localhost:11434
claude --model qwen3.5:4b
localhost:11434 is the address of Ollama's local server.
If you have questions or try another model, let me know in the comments!
