Hostinger VPS DeepSeek Deployment Guide: Complete Installation and Optimization Tutorial

📢公告：欢迎加入本站TG群，每月1-2次抽免费VPS

With the rapid development of artificial intelligence, DeepSeek has become a popular AI model for developers and businesses, especially for foreign trade websites, cross-border e-commerce, and automated content generation. For users running services on Hostinger VPS, deploying DeepSeek efficiently can greatly improve productivity. This guide walks through the complete process from installing Ollama to optimizing DeepSeek performance.

1. Requirements for Running DeepSeek on Hostinger VPS

• CPU: 4 cores or higher
• RAM: 16GB or more
• Storage: 12GB+ available space
• OS: Ubuntu / Debian / CentOS / RockyLinux

Recommended Hostinger VPS Plans

2. Install Ollama (Required for DeepSeek)

1. Update system sudo apt update
2. Install dependencies sudo apt install python3 python3-pip python3-venv
3. Install Ollama curl -fsSL https://ollama.com/install.sh | sh
4. Create virtual environment python3 -m venv ~/ollama-webui && source ~/ollama-webui/bin/activate
5. Install Web UI pip install open-webui
6. Run screen session screen -S ollama
7. Start Web UI open-webui serve
8. Open browser http://YOUR_IP:8080

3. Install DeepSeek Model

1. Download model ollama run deepseek-r1:7b
2. Wait for download (~5GB)
3. Check installed models ollama list
4. Restart UI open-webui serve

4. Test DeepSeek

1. Open WebUI
2. Create account
3. Select DeepSeek model
4. Send test prompt

5. Optimize DeepSeek Performance

• Streaming: faster output, more CPU
• Temperature: creativity level
• Reasoning: accuracy level
• Max tokens: response length

Example Settings

Temperature: 0.2
Reasoning: 80
Max tokens: 512

6. Auto Start DeepSeek with systemd

nano /etc/systemd/system/deepseek.service

[Unit]
Description=DeepSeek AI
After=network.target

[Service]
ExecStart=/root/ollama-webui/bin/open-webui serve
Restart=always

[Install]
WantedBy=multi-user.target

systemctl enable deepseek
systemctl start deepseek

Summary