Install Qwen 3.5 (0.8B) Locally with Ollama

Need a VPS? Get one at Hostinger - use coupon code HOWTO20 for 20% off.

What Is Qwen 3.5?

Alibaba's Qwen team released four small models built on the same Qwen 3.5 foundation as their flagship 397B model. The 0.8B is the smallest in the series, only about 1 GB to download, and runs on any machine including CPU-only servers. The whole series ships under Apache 2.0 so it's fully open and free to use commercially.

Ollama library page: https://ollama.com/library/qwen3.5

Prerequisites

A VPS or any machine running Linux/macOS/Windows - get a VPS at Hostinger with 20% off using code HOWTO20
No GPU required
At least 2 GB of RAM

Step 1: Install Ollama

Run the official Ollama install script:

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

Verify it's installed:

ollama --version

Step 2: Pull the Model

This downloads the Qwen 3.5 0.8B model. It's about 1 GB.

ollama pull qwen3.5:0.8b

Step 3: Run the Model

Start a chat session directly in your terminal:

ollama run qwen3.5:0.8b

Type any prompt and press Enter. Press Ctrl+D to exit.

Useful Links

Qwen 3.5 on Ollama: https://ollama.com/library/qwen3.5
Qwen 3.5 GitHub: https://github.com/QwenLM/Qwen3.5
HuggingFace Models: https://huggingface.co/collections/Qwen/qwen35