I am a programmer of many years and I consider myself quite skilled in that area. An artist I am not. Often I ran into the issue that for game programming, I am stuck with ugly developer art. Because, well, an artist, I am not.

For quite a while I've been trying to get AI to make me concept art for my characters. And even when that worked, I never, ever got the poses right. Until now. Until Qwen Image Edit 2509. I watched many a tutorial, read lots of stuff, and based on all that, I finally have a working workflow that does one thing:

It takes a character, it takes a pose, and it poses that character in that pose

I am sharing this workflow here. Hopefully others find it useful as well. Or have better ways to do this.

Resources you might need

For hardware, I have a 3080-10GB and 64GB system RAM on a Ryzen 9 5900X 12-Core. Running Linux Mint. This works fine, about 70 seconds per image generated.

I am using the GUFF models of Qwen 2509
https://huggingface.co/QuantStack/Qwen-Image-Edit-2509-GGUF/tree/main
Most of the other things needed are here
https://docs.comfy.org/tutorials/image/qwen/qwen-image-edit

And Aldea Lab made a pretty useful Youtube video about it here