How to Install Mistral on Ubuntu 24.04

Mistral is a cutting-edge open-weight large language model (LLM) developed by Mistral AI, a European AI research company. It is designed to compete with models like GPT-4, LLaMA, and DeepSeek by offering high-performance natural language processing (NLP), multilingual support, and open-source accessibility.

Key Features of Mistral AI

Open-Weight Large Language Model
- Provides freely accessible model weights, allowing developers to run, modify, and fine-tune the model locally.
Optimized for Performance and Efficiency
- Uses Mixture of Experts (MoE) architecture in some versions to reduce computational costs while maintaining high accuracy.
Multilingual Capabilities
- Supports multiple languages, making it useful for global AI applications.
High-Quality Text Generation
- Capable of writing, summarizing, translating, and coding with near-human fluency.
Fine-Tuning and Customization
- Allows businesses and researchers to train the model on domain-specific data for enhanced AI capabilities.
On-Premises Deployment
- Unlike proprietary models like GPT-4, Mistral can run locally, providing better control over data privacy and security.
API Access for Developers
- Offers API-based access for seamless integration into AI-powered applications.
Open-Source Availability
- Certain Mistral models are fully open-weight, enabling research and transparency in AI development.

Advantages of Mistral AI

Privacy and Control: Runs locally or on-premises, unlike cloud-based LLMs that require third-party servers.
Faster and Cost-Efficient: Uses MoE techniques to optimize computing resources.
Competitive with GPT-4 & LLaMA: Offers similar performance levels in NLP tasks.
Flexible Deployment: Can be fine-tuned for specific industries such as finance, healthcare, and legal AI applications.
Scalable for AI-Powered Applications: Useful for chatbots, virtual assistants, and automated workflows.

Use Cases for Mistral AI

AI Chatbots and Virtual Assistants
- Used for customer service, AI-driven support, and automation.
Advanced Text Processing
- Capable of summarizing, paraphrasing, and translating text efficiently.
Code Generation and Assistance
- Helps developers write, debug, and optimize code in multiple programming languages.
AI-Powered Content Creation
- Assists in article writing, storytelling, and SEO content.
Enterprise AI Solutions
- Used by businesses for document analysis, research assistance, and workflow automation.
Scientific and Academic Research
- Enables AI-driven insights for technical writing, machine learning experiments, and data analysis.

Mistral vs. Other Large Language Models

Feature	Mistral	GPT-4	LLaMA	DeepSeek
Open-Source	✅ Yes	❌ No	✅ Yes	✅ Yes
Multilingual	✅ Yes	✅ Yes	⚠️ Limited	✅ Yes
Fine-Tuning	✅ Yes	✅ Yes	✅ Yes	✅ Yes
Code Generation	✅ Yes	✅ Yes	✅ Yes	✅ Yes
Local Deployment	✅ Yes	❌ No	✅ Yes	✅ Yes

Mistral provides a powerful alternative to proprietary models like GPT-4 while maintaining openness, efficiency, and high-performance NLP capabilities.

Why Choose Mistral AI?

Mistral is a flexible, high-performance LLM that balances efficiency, privacy, and open-source accessibility. It is ideal for researchers, developers, and businesses looking to integrate advanced AI into their workflows without relying on proprietary cloud-based models.

Step 1: Set Up a Server Instance

To begin, you need an Ubuntu 24.04 server instance. Follow these steps to create one:

Access Shape.Host Dashboard: Log in to your Shape.Host account and navigate to the Dashboard.

Click “Create”: Look for the “Create” button at the top-right corner and click it.

Select Server Type: Choose “Instances” to start configuring your server environment.

Choose a Data Center Location: Pick a location nearest to your audience for optimal performance.

Shape.host instance form with the data center location selector highlighted.

Select a Hosting Plan: Choose a plan that aligns with your project requirements, whether Standard or Memory-Optimized.

Set Up the Operating System: Select Ubuntu 24.04 as the server’s OS.

Shape.host instance form with the operating system image selector highlighted.

Finalize Configuration: Select your preferred authentication method (SSH keys or password) and click Create Instance to launch the server.

Shape.host instance form with the Create Instance button highlighted.

In Dashbord you will find your Instance IP

Shape.host Resources dashboard with the server IP address highlighted.

Step 2: Connect to Your Instance

Once your server is running, connect to it using SSH:

Linux/macOS: ssh root@<your_server_ip>
Windows: Use PuTTY. Enter your server’s IP, select SSH, and log in with your credentials.

Step 3: Update System and Install Required Packages

Step 3.1: Update the System

First, update and upgrade all system packages:

apt update && apt upgrade -y

Terminal showing APT package management commands and output.

Step 3.2: Install Python and Pip

Install Python 3 and Pip, which are required for Mistral’s dependencies:

apt install python3  
apt install python3-pip

Step 3.3: Install Git

Git is required for managing repositories:

apt install git

Step 4: Set Up the Working Directory

Create a directory for the project:

cd /home/  
mkdir webui  
cd webui

Step 5: Install and Configure Ollama

Step 5.1: Install Ollama

Ollama is required to manage and run Mistral models:

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

Terminal showing file and directory setup commands.

Step 5.2: Verify Ollama Service

Check if the Ollama service is running correctly:

systemctl status ollama.service

Terminal showing systemd service management commands or output.

Step 5.3: Pull Mistral Models

Download different versions of the Mistral model:

ollama pull mistal

Terminal showing Ollama model download progress.

Step 5.4: List Available Models

Confirm that the models were successfully downloaded:

ollama list

Terminal displaying the list of locally available Ollama models.

Step 5.5: Run Mistral

To test Mistral, run the following command:

ollama run mistral

Step 6: Set Up Python Virtual Environment

Step 6.1: Install Python Virtual Environment

apt install python3-venv

Step 6.2: Create and Activate the Virtual Environment

python3 -m venv /home/webui/open-webui-venv  
source /home/webui/open-webui-venv/bin/activate

Step 6.3: Install Open WebUI

pip install open-webui

Terminal showing Django or Python virtual environment commands.

Step 7: Configure Open WebUI as a System Service

Step 7.1: Create a Systemd Service File

Open the service configuration file for editing:

nano /etc/systemd/system/open-webui.service

(Insert the required service configuration inside the file, as per project requirements.)

[Unit]
Description=Open WebUI Service
After=network.target

[Service]
User=root
WorkingDirectory=/home/webui/open-webui-venv
ExecStart=/home/webui/open-webui-venv/bin/open-webui serve
Restart=always
Environment="PATH=/home/webui/open-webui-venv/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"

[Install]
WantedBy=multi-user.target

Terminal showing a systemd service unit configuration.

Step 7.2: Reload Systemd Daemon

systemctl daemon-reload

Step 7.3: Enable and Start Open WebUI Service

systemctl enable open-webui.service  
systemctl start open-webui.service

Step 8: Access Open WebUI

After setting up Open WebUI, you can access it via your web browser.

Find Your Server’s IP Address

To check your server’s IP address, run:

hostname -I

Open the Web UI in Your Browser

Once the Open WebUI service is running, open a web browser and enter:

http://<your_server_ip>:8080

Replace <your_server_ip> with the actual IP address of your server.

Now, you should be able to interact with Mistral via Open WebUI.

Open WebUI initial administrator account form.

You have successfully installed and configured Mistral on Ubuntu 24.04. Your system is now ready to use Ollama and Open WebUI for interacting with AI models.

For scalable and high-performance hosting, consider Shape.Host Linux SSD VPS, optimized for AI and machine learning workloads.