Wan 2.2
Generate videos from text or an image – one of the best open video models.
DifficultyAdvanced
HardwareNVIDIA GPU with 8 GB VRAM or more, ideally 24 GB (e.g. RTX 3090/4090), 32 GB RAM, about 17 GB of disk space
LicenseApache-2.0
Wan 2.2 generates short video clips (5 seconds, 1280×704). We use the 5B variant (TI2V) with the official ComfyUI template. Thanks to automatic offloading, it runs with as little as 8 GB VRAM, and much faster with 24 GB. Prerequisite: ComfyUI is installed (see ComfyUI in the Toolbox). Depending on your graphics card, a clip takes roughly 5 to 30 minutes.
How to try it
Commands from the official guides – just copy them.
Linux
# Prerequisite: ComfyUI installed following our guide (folder ~/ki/ComfyUI)
# 1. Download the three Wan 2.2 models (about 17 GB in total):
cd ~/ki/ComfyUI/models
wget -c -P diffusion_models https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/resolve/main/split_files/diffusion_models/wan2.2_ti2v_5B_fp16.safetensors
wget -c -P text_encoders https://huggingface.co/Comfy-Org/Wan_2.1_ComfyUI_repackaged/resolve/main/split_files/text_encoders/umt5_xxl_fp8_e4m3fn_scaled.safetensors
wget -c -P vae https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/resolve/main/split_files/vae/wan2.2_vae.safetensors
# 2. Download the ready-made workflow:
wget -O ~/ki/wan22-video.json https://raw.githubusercontent.com/Comfy-Org/workflow_templates/main/templates/video_wan2_2_5B_ti2v.json
# 3. Start ComfyUI:
source ~/ki/comfy-env/bin/activate && comfy --workspace ~/ki/ComfyUI launch
# 4. Open http://127.0.0.1:8188 in your browser and drag the file ~/ki/wan22-video.json into the window.
# Enter your prompt in the "CLIP Text Encode (Positive Prompt)" field (English works best),
# then click "Run" at the top. The video ends up in ~/ki/ComfyUI/output/video/macOS
# Runs on Apple Silicon (with 32 GB of memory or more), but very slowly – a clip can take over an hour.
# Prerequisite: ComfyUI installed following our Terminal guide (folder ~/ki/ComfyUI)
cd ~/ki/ComfyUI/models
curl -L -C - --create-dirs -o diffusion_models/wan2.2_ti2v_5B_fp16.safetensors https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/resolve/main/split_files/diffusion_models/wan2.2_ti2v_5B_fp16.safetensors
curl -L -C - --create-dirs -o text_encoders/umt5_xxl_fp8_e4m3fn_scaled.safetensors https://huggingface.co/Comfy-Org/Wan_2.1_ComfyUI_repackaged/resolve/main/split_files/text_encoders/umt5_xxl_fp8_e4m3fn_scaled.safetensors
curl -L -C - --create-dirs -o vae/wan2.2_vae.safetensors https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/resolve/main/split_files/vae/wan2.2_vae.safetensors
curl -L -o ~/ki/wan22-video.json https://raw.githubusercontent.com/Comfy-Org/workflow_templates/main/templates/video_wan2_2_5B_ti2v.json
source ~/ki/comfy-env/bin/activate && comfy --workspace ~/ki/ComfyUI launch
# In your browser at http://127.0.0.1:8188, drag the file ~/ki/wan22-video.json into the window, change the prompt, click "Run".Windows (PowerShell)
# Prerequisite: ComfyUI Desktop installed (https://www.comfy.org/download), default folder Documents\ComfyUI
# 1. In PowerShell, download the three models (about 17 GB):
cd "$([Environment]::GetFolderPath('MyDocuments'))\ComfyUI\models"
curl.exe -L -C - --create-dirs -o diffusion_models\wan2.2_ti2v_5B_fp16.safetensors https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/resolve/main/split_files/diffusion_models/wan2.2_ti2v_5B_fp16.safetensors
curl.exe -L -C - --create-dirs -o text_encoders\umt5_xxl_fp8_e4m3fn_scaled.safetensors https://huggingface.co/Comfy-Org/Wan_2.1_ComfyUI_repackaged/resolve/main/split_files/text_encoders/umt5_xxl_fp8_e4m3fn_scaled.safetensors
curl.exe -L -C - --create-dirs -o vae\wan2.2_vae.safetensors https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/resolve/main/split_files/vae/wan2.2_vae.safetensors
# 2. Download the ready-made workflow to your Downloads folder:
curl.exe -L -o "$HOME\Downloads\wan22-video.json" https://raw.githubusercontent.com/Comfy-Org/workflow_templates/main/templates/video_wan2_2_5B_ti2v.json
# 3. Start ComfyUI, drag the file wan22-video.json from "Downloads" into the window,
# enter your prompt in the "CLIP Text Encode (Positive Prompt)" field and click "Run".Ask the AI assistant
Questions about Wan 2.2, installation or background? Our assistant knows this page and runs on our own machine, not in someone else's cloud.
The Assistant is an AI (Gemma 4) and can make mistakes – please check important details against the source. Questions are used only to generate an answer and are deleted after one hour. Please don't enter any personal data.
Comments
What do you think? Already tried it? Write it here, no sign-up needed.
Comments are public. Links aren't allowed, and we delete abusive posts. You can delete your own comment at any time in this browser. Learn more in the Privacy Policy.