Installing Ollama.

Or: Using an LLM Manager.


5 min read

Installing Ollama.

Published: Saturday 12th December 2023.
Update: Sunday 31st December 2023 (yes, I'm working on New Year's Eve.)
Update: Tuesday 20th February 2024.


This post describes the installation of Ollama, a local large language model (LLM) manager. It requires a Linux-based distro and Anaconda. After setting up an environment in Anaconda, Ollama is installed and used to download and run LLMs. I use the Mistral model as an example. Ollama enables the use of powerful LLMs for research, development, business, and personal use.


An Introduction.

LLMs (large language models) are amazing machines. They are the bleeding edge of modern technology and devs should learn, and adapt to, these awesome engines.

The purpose of this post is to describe how to install Ollama, a local LLM manager.

The Big Picture.

There are many open-source LLMs. Ollama provides easy access to a small set of those models. Visit the Ollama models library to get a list of the LLMs they support. The index is continually updated, so I frequently revisit this archive.

I love Hugging Face, but it's also nice to have a curated series of models.


  • A Linux-based distro (I use Ubuntu), and

  • Anaconda.

What is Anaconda?

Python projects can run in virtual environments, which are isolated spaces for managing dependencies, where different versions of the same package can run in different environments while avoiding version conflicts; venv is a built-in Python 3.3+ module for running virtual environments, Anaconda is a Python and R distribution for scientific computing that includes the conda package manager, and Miniconda is a small bootstrap version of Anaconda that includes only conda, Python, packages they both depend on, and a small number of other useful packages (like pip and zlib).

Using Anaconda to Setup an Environment.

  • I update my system:
sudo apt clean && \
sudo apt update && \
sudo apt dist-upgrade -y && \
sudo apt --fix-broken install && \
sudo apt autoclean && \
sudo apt autoremove -y
  • I use conda to display a list of Anaconda environments:
conda env list
  • I use conda to create, and activate, a new environment named (-n) (ollama):
conda create -n ollama python=3.11 -y && conda activate ollama

NOTE: This command creates the (ollama) environment, then activates the (ollama) environment.

Changing the env Home Directory.

NOTE: I will define the home directory with settings in the environment directory.

  • I create the Ollama home directory:
mkdir ~/Ollama
  • I make new directories within the (ollama) environment:
mkdir -p $HOME/anaconda3/envs/ollama/etc/conda/activate.d
  • I use the Nano text editor to create the shell script:
sudo nano ~/anaconda3/envs/ollama/etc/conda/activate.d/
  • I copy the following, add it (CTRL + SHIFT + V) to the script, save (CTRL + S) the changes, and exit (CTRL + X) Nano:
cd ~/Ollama
  • I activate the (base) environment:
conda activate
  • I activate the (ollama) environment:
conda activate ollama

NOTE: I should now, by default, be in the ~/Ollama home directory.

What is Ollama?

Ollama is a tool that is used to download, set up, and run large language models on a local PC. It lets me use powerful models like Llama 2 and Mistral on my personal computer. Ollama natively runs on Linux, macOS, and Windows.

Installing Ollama.

  • I install Ollama:
curl | sh
  • I list the LLMs downloaded by Ollama:
ollama list
  • If the above command fails, I run Ollama as a background service:
ollama serve &
  • If the following error shows when running the previous command, that means Ollama is already running as a background service:
Error: listen tcp bind: address already in use

Using Ollama to Run the Mistral Model.

  • I run the Mistral model:
ollama run mistral

NOTE 1: The ollama run command performs an ollama pull if the model has not already been downloaded.

NOTE 2: The ollama run command is used to run the named LLM.

NOTE 3: I sometimes need to restart my system to properly run the LLM.

Testing the Mistral Model.

  • At the >>> prompt, I run the following command:
tell me a joke

NOTE: Do NOT expect the joke to be any good.

The Results.

By following the steps outlined in this post, I can set up dedicated Anaconda environments, install Ollama, download LLMs from, and test them locally. This process enables me to evaluate the power of cutting-edge models for research, development, business, and personal use.

NOTE: Commercial use depends on the licenses of each model.

In Conclusion.

LLMs are the bleeding edge of modern technology and we should adapt to these awesome engines.

I've been exploring Ollama, a local LLM manager that provides easy access to a selection of open-source LLMs from Their list of supported LLMs is continually updated. It's a buffet of cutting-edge tech that keeps on growing!

The installation process was straightforward because all I needed was a Linux-based distro and Anaconda. I created and activated a new environment named (ollama) using the conda command. Then, I set up an Ollama home directory and made new directories within the (ollama) environment. Once I got my environment set up, I installed Ollama and started downloading and running an LLM. I've been testing it out with the 'Mistral' model. And let me tell you, it's impressive!

By following these steps, I can set up dedicated Anaconda environments, install Ollama, download LLMs from, and test them locally.

This is a game-changer for research, development, business, and personal use. Future Tech is here and it's time to embrace it.

Have you used Ollama? What has been your experience? Let's discuss in the comments below!

Until next time: Be safe, be kind, be awesome.