How to Deploy WAN 2.1 on GCP for AI Video Workflows

For effective workflows, Alibaba's open-source Wan 2.1 video model thrives on Google Cloud Platform (GCP). This guide describes how technical people can utilize Google's AI infrastructure to install it on GCP using Compute Engine as of March 13, 2025.

• GCP Account: With Compute Engine enabled.
• Hardware: A2-highgpu-1g (A100, 40GB VRAM).

Step 1: Create VM

• Machine Type: A2-highgpu-1g.
• Boot Disk: Ubuntu 20.04, 50GB.

Step 2: Connect

SSH:

gcloud compute ssh wan2-1-vm --zone us-central1-a

Step 3: Install and Download

Setup:

sudo apt update && sudo apt install -y python3-pip git
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

Execute:

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

• Instance: A2-highgpu-1g
• VRAM: 40GB
• Cost/Hour: $3.674

Wan 2.1 on GCP’s A2 instances delivers high performance video generation as of March 2025. Ideal for developers, this setup harnesses Google’s GPU power, ensuring efficient AI video workflows.

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