How to Deploy WAN 2.1 on AWS EC2 - A Complete Guide

Alibaba released Wan 2.1's T2V-14B model on February 25, 2025, and it needs strong computation to produce high quality videos. Targeting developers in need of scalable solutions, this guide describes its implementation on Amazon Web Services (AWS) utilizing EC2 as of March 13, 2025.

• AWS Account: With EC2 access.
• Hardware: g5.xlarge (NVIDIA A10G, 24GB VRAM).

 

Step 1: Launch EC2

AMI: Deep Learning AMI (Ubuntu 20.04).

Instance: g5.xlarge.

Storage: 50GB.

Step 2: Connect

SSH:

ssh -i <key.pem> ubuntu@<ec2-ip>

Step 3: Install and Download

Clone and download:

git clone https://github.com/Wan-Video/Wan2.1.git
cd Wan2.1
huggingface-cli download Wan-AI/Wan2.1-T2V-14B --local-dir ./models
pip3 install -r requirements.txt

Step 4: Run Wan2.1

Generate video:

python3 inference.py --model_path ./models/Wan2.1-T2V-14B --prompt "A futuristic city" --output city.mp4

• Instance: g5.xlarge
• VRAM: 24GB
• Cost/Hour: $1.006

As of March 2025, scalable video creation is made possible by deploying Wan 2.1 on AWS with a g5.xlarge instance. This configuration balances cost and performance for sophisticated AI video operations, making it ideal for companies and developers.

Ready to transform your business with our technology solutions? Contact Us  today to Leverage Our AI/ML Expertise.