Maj Ollama

Are you using Ollama on Windows to run local artificial intelligence models (like Qwen or Mistral), and have you noticed an update notification?

No panic! This simple, beginner-friendly guide walks you through updating Ollama quickly via PowerShell, without risking the deletion of your already downloaded models.

Why update Ollama?

Regularly updating Ollama allows you to:

  • Optimize performance: benefit from the latest inference engine improvements (llama.cpp) for faster responses.
  • Improve stability: reduce RAM/VRAM memory usage on long prompts.
  • Access new models: support the latest AI model architectures.

Step 1: Exit the current interface

If you are stuck on a text menu or an interface in your terminal (with options like Copilot CLI or Claude Code), start by cleanly exiting this screen:

  1. Press the Esc key on your keyboard to close the interactive interface.

Step 2: Open Windows PowerShell

We will run the official update command in PowerShell:

  1. Press the Windows key on your keyboard.
  2. Type PowerShell.
  3. Click on Windows PowerShell to open it.

Step 3: Run the update command

In the PowerShell window, copy and paste the official command below, then press Enter:

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

What does this command do?

  • irm (Invoke-RestMethod): downloads the official Ollama installation script.
  • | iex (Invoke-Expression): safely executes the script to replace the old version with the new one.

The script will download and install the latest version. You should get a result similar to this:

text 
>>> Downloading Ollama for Windows...
######################################## 100.0%
>>> Installing Ollama...
>>> Install complete. Run 'ollama' from the command line.

Step 4: Verify the update and your models

Once the installation is complete, verify that everything went well.

1. Check the installed version

Type the following command in PowerShell:

powershell 
ollama --version

Example return: ollama version is 0.40.12

2. Check your existing models

To ensure that your models (qwen2.5, mistral-nemo, etc.) are still present and intact, run:

powershell 
ollama list

Your models will appear as a list with their size and modification date.

FAQ: Is there a risk my models will be deleted?

No, absolutely not.

The Ollama program is stored in your system files (AppData\Local\Programs\Ollama), while your downloaded models are kept separately in your user folder (.ollama\models) or on your external drive. The update only replaces the Ollama software, not your data.

Cet article est rédigé par Blaise Grenier, Journaliste reporter d'images et auteur. Découvrir le parcours de l'auteur et la charte éditoriale »