from 24 GBImages
Photo editor · Qwen Edit
Edit a photo with words: swap the background, turn day into evening, remove an object.
from €0.30/h · RTX 3090, 24 GB
Ready-made app · Qwen-Image
Run Qwen-Image on a rented GPU
Alibaba’s Qwen-Image draws pictures with lettering in them: posters, signs, covers and banners. It handles accented and non-Latin characters too. The ready-made app needs a card with 24 GB, and the server installs ComfyUI and the model on its own.
- App
- Images · Qwen-Image
- GPU
- RTX 3090, 24 GB
- Price
- from €0.30/h
To launch, your balance must cover one hour of the machine. You are billed by the minute for actual use. The server installs Images · Qwen-Image for you.
from 24 GBof video memory
2–5 minutesuntil the server is up
By the minutebilled from your balance
ComfyUIopens in your browser
What is on the server after launch
Images · Qwen-Image is free: you only pay for GPU time. Everything installs at launch; there is nothing to configure by hand.
- ComfyUI in your browser
- Qwen-Image in FP8, Apache 2.0 licence
- Qwen2.5-VL 7B text encoder and VAE
- A button in the dashboard that opens ComfyUI
- Your folder with models and results can be saved when you stop
Licence Apache 2.0: you can use the images in commercial projects.
How much memory you need and which GPU to pick
Images · Qwen-Image needs a card with at least 24 GB; the cheapest suitable one is the RTX 3090. 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 |
|---|---|---|---|
| Qwen-Image in FP8, ready-made app | from 24 GB | RTX 3090 | €0.30 |
| Qwen-Image without quantisation | from 48 GB | L40 | €0.46 |
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 Qwen-Image 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 Qwen-Image. The dashboard opens with Images · Qwen-Image and a card with at least 24 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
Qwen-Image in an empty ComfyUI
cd ComfyUI/models
wget -P diffusion_models https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/split_files/diffusion_models/qwen_image_2512_fp8_e4m3fn.safetensors
wget -P text_encoders https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors
wget -P vae https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/split_files/vae/qwen_image_vae.safetensors
# workflow: Workflow → Browse Templates → Qwen-Image
What people run Qwen-Image for
01
Posters and flyers
Title, date and tagline right in the image, no editing afterwards.
02
Covers and banners
Video thumbnails, article covers and ad creatives with text.
03
Signs and menus
A mock-up of a sign, a label or a menu board to show a client the idea.
Tips before you launch
- Put the lettering in quotes right in the prompt: the model will carry it into the image.
- Short lettering comes out cleaner. In a long phrase the model sometimes gets a letter wrong; another seed usually fixes it.
- The app includes a 4-step LoRA: drafts render several times faster; re-render the final one with full steps.
Other ready-made apps
Each app installs its interface and model on its own. You only pay for GPU time.
from 16 GBImages
Images · Flux
Images from a text prompt: illustrations, covers, social posts and product shots.
from €0.16/h · RTX 4060 Ti, 16 GB
from 24 GBImages
Images · HiDream-O1
Images from a prompt with a fresh open model. The MIT licence allows commercial use.
from €0.30/h · RTX 3090, 24 GB
Frequently asked questions
Which GPU do I need for Qwen-Image?
The ready-made FP8 app needs a 24 GB card: RTX 3090, RTX 4090 or RTX 5090. Unquantised, the model takes about 40 GB and needs a 48 GB card.
Can Qwen-Image write in languages other than English?
Yes. Short lettering comes out clean; in long phrases a letter is occasionally wrong. Another seed or shorter text helps.
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.
A server is usually ready in 2–5 minutes
Sign up in a minute, top up from €10, pay by the minute. Stop the server and billing stops.