# I ran Doom inside the Qwen model weights! — retained record

> Retained reddit record from reddit:LocalLLaMA.

Canonical: https://highsignal.app/data/records/18f3a384f828e27a

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:LocalLLaMA)  published 2026-10-28 22:24 ingested 2026-10-29 02:42

[original source ↗](https://www.reddit.com/r/LocalLLaMA/comments/1xi2jfm/i_ran_doom_inside_the_qwen_model_weights/)

## retained text



```
Inspired by the [qwenputer](https://www.reddit.com/r/LocalLLaMA/comments/1xhmo1q/someone_built_a_32bit_riscv_cpu_directly_inside/) project posted in this sub earlier today, I managed to get a basic version of Doom running on the virtual machine created in the post.

  
It's quite slow, but here we are. Doom running on an LLM.

  
Project repo: [https://github.com/mapestree/qwenputer-doom](https://github.com/mapestree/qwenputer-doom)
```



document fetched 2026-10-29 00:44 · canonical https://reddit.com/r/LocalLLaMA/comments/1xi2jfm/i_ran_doom_inside_the_qwen_model_weights
