Qwen-Image: pictures where the text is readable
Image generation with best-in-class text rendering — Apache 2.0
- Video memory
- 16 GB
- Setup
- ~25 min
- Access
- port 8188
- Billing
- hourly, in BRL
Anyone making campaign art with AI knows the problem: the picture looks great and the words come out scrambled. Qwen-Image is the open model that handles this best today. Here it arrives with all three files downloaded and ComfyUI running.
Current-generation image model with all three files preloaded (model, text encoder and VAE) and ComfyUI ready to use. Best-in-class at rendering TEXT inside the image. Apache 2.0: commercial use of the output is allowed.
What it is for
- Ad creatives with readable headlines and prices, no retouching
- Social posts in series, same layout, different copy
- Apache 2.0: sell what you generate, no permission needed
- Repeatable ComfyUI workflows, stored on your own machine
How to deploy
- Create your account and add balance (card or Pix, no subscription).
- In the console, pick the Qwen-Image (ComfyUI) template and a machine — the console hides the ones that do not meet the requirement.
- In about 25 minutes the setup finishes and the access address shows up in the panel, on port 8188.
Done? Just destroy the machine and billing stops with it. No contract, no minimum commitment.
FAQ
How is it different from FLUX.2?
Image quality is comparable, the license is not: FLUX.2 open weights are non-commercial — selling the output requires a paid license from the vendor. Qwen-Image is Apache 2.0, which is why it is what we offer in one click.
How much video memory do I need?
16 GB runs the quantised build we install. A 24 GB card gives more headroom at high resolution.
Do the models come preloaded?
Yes — the template downloads model, text encoder and VAE before opening the interface, precisely so you do not click generate and hit a missing-node error.