Ready-made app · Video generation

AI video generation on a rented GPU: Wan 2.2

Video generation needs a card with at least 24 GB: that is where the fast Wan 2.2 5B runs. The full Wan 2.2 14B in the ready-made app needs 48 GB, but its clips are noticeably better and it can animate photos. According to the developers, Wan 2.2 5B renders 5 seconds of 720p in about 9 minutes on a 24 GB card.

App: Video · Wan 2.2 14B
GPU: L40, 48 GB
Price: from €0.46/h

Launch Wan 2.2 14B · from €0.46/h Other servers

To launch, your balance must cover one hour of the machine. You are billed by the minute for actual use. The server installs Video · Wan 2.2 14B for you.

Glass boxes with a chat bubble, a picture, a microphone and a film strip on a conveyor belt

from 48 GBof video memory

2–5 minutesuntil the server is up

By the minutebilled from your balance

ComfyUIopens in your browser

Examples: what went in and what came out

Prompts and results from the ready-made app, no retouching. You can run the same prompt on your own server.

Wan 2.2 14B

Prompt · EN

Aerial drone shot flying over snowy mountains at sunrise, clouds below, cinematic

5 s, 640×640 · 13 min on an RTX A6000

Launch Wan 2.2 14B · from €0.46/h L40, 48 GB · billed by the minute

Wan 2.2 5B

Prompt · EN

A golden retriever running on a beach at sunset, waves splashing, slow motion, cinematic

5 s, 1280×704 · 3.5 min on an RTX 4090

Launch Wan 2.2 5B · from €0.30/h RTX 3090, 24 GB · billed by the minute

What is on the server after launch

There are two ready-made apps. Both install ComfyUI and download their model: Wan 2.2 14B gives higher quality, Wan 2.2 5B runs on a cheaper card.

Video · Wan 2.2 14B

from 48 GB

High-quality clips from text or from a photo.

ComfyUI in your browser
Wan 2.2 14B in FP8: text to video and image to video
4-step LoRA: clips render several times faster

from €0.46/h · L40 Launch Wan 2.2 14B

Video · Wan 2.2 5B

from 24 GB

A quick draft of a clip on an affordable card, to test the idea before the final version.

ComfyUI in your browser
Wan 2.2 TI2V 5B
UMT5 text encoder and Wan 2.2 VAE

from €0.30/h · RTX 3090 Launch Wan 2.2 5B

How much memory you need and which GPU to pick

Video · Wan 2.2 14B needs a card with at least 48 GB; the cheapest suitable one is the L40. The table shows memory for other versions and tasks. Prices update from available catalogue machines when you open the page.

Version	Video memoryMemory	Suitable GPUGPU	Price per hourPer hour
Wan 2.2 5B, 720p	from 24 GB	RTX 3090	€0.30
Wan 2.2 14B in FP8, ready-made app	from 48 GB	L40	€0.46
Wan 2.2 14B without quantisation	from 80 GB	A100	€1.46

Guide from real launches: complex graphs and high resolutions need more memory. Pick: the cheapest single card with enough memory. All GPUs and prices

How to run video generation in 3 steps

Sign up and top up. Create an account and top up from €10. To launch, your balance must cover at least one hour of the chosen machine; after that you pay by the minute.
Launch the app. Click Launch Wan 2.2 14B. The dashboard opens with Video · Wan 2.2 14B and a card with at least 48 GB already selected; check the price and confirm.
Work in your browser. The server starts in 2–5 minutes, and the app downloads the model in a few more. Then press the ComfyUI button in the dashboard and type your prompt.

Finished? Stop the server in the dashboard and billing stops. Save and stop keeps /workspace for €0.03 per GB per month, and it comes back on your next server automatically.

Commands for a manual install on an empty server

ComfyUI with Wan 2.2 workflows

git clone https://github.com/comfyanonymous/ComfyUI && cd ComfyUI pip install -r requirements.txt python main.py --listen 0.0.0.0 --port 8188 # Wan 2.2 workflows: Workflow → Browse Templates → Video

What people run video generation for

A clip from a description

Describe the scene in words and the model makes a clip a few seconds long.

Bring a photo to life

Wan 2.2 14B turns a still into a short video with camera and subject movement.

Social and ad inserts

Short scenes for Reels, Shorts and ad creatives without a shoot.

Draft on 5B, final on 14B

Test the idea on an affordable card and re-render the winning prompt in high quality.

Tips before you launch

An RTX 4090 or 5090 is enough for single clips and experiments; an H100 makes sense when the model runs unquantised or a series has to render fast.
Start at 480p: it is 2–3 times faster, and you can re-render a good seed in 720p later.
Downloading weights is billed as server time, so prepare your list of models in advance.

Frequently asked questions

Which GPU do I need for AI video generation?

The ready-made Wan 2.2 14B app needs a 48 GB card (RTX A6000, L40S), and the fast Wan 2.2 5B is fine on 24 GB (RTX 3090, RTX 4090).

How is Wan 2.2 14B different from 5B?

14B makes noticeably better clips and can turn a photo into a video, but it needs a 48 GB card. 5B is faster and cheaper: test the idea on it, then render the final version on 14B.

How long does a video take on an RTX 4090?

According to the developers, Wan 2.2 TI2V-5B makes a 5-second 720p clip in under 9 minutes on a single 24 GB card. Heavier models are slower.

Can I use the clips commercially?

It depends on the model’s licence. Wan 2.2 is released under Apache 2.0; other video models have their own licences, some with restrictions. Check the terms before you publish.

How long does a launch take?

The server usually starts in 2–5 minutes. Then the app downloads the model on its own, which takes a few more minutes, and a button that opens ComfyUI appears in the dashboard.

What happens to my models and results after I stop?

When you stop, click Save and stop: the /workspace folder with your models, LoRAs and results goes to storage for €0.03 per GB per month and comes back on your next server automatically. If you stop without saving, the disk is deleted.

How is it billed?

By the minute from your balance in euros. Top up from €10; to launch, your balance must cover at least one hour of the chosen machine. Stop the server and billing stops.