Tencent-Hunyuan/Hy-Nomas ? reverse-engineered prompt

Reverse engineered prompt

Build me a web app for free form interactive world generation where I can start with either text or an uploaded image, then keep chatting with it over multiple turns to change the scene while also moving the camera around. I want it to feel like exploring a living generated world, with the history of what I said still affecting what happens next.

It should run in a low memory setup on a single RTX 4090 if possible, and I’d like a simple local demo I can open in the browser. Please include a clean way to try the model, control the camera, and save or resume a session. If there’s a good way to support LoRA fine tuning for downstream use, add that too.

Use the current docs online if you need to, and make the setup as easy as you can so I can get it running quickly.

Are you gonna build this?

make sure you review the code using coderabbit

Try freeSponsored — opens CodeRabbit in a new tab