# Is Qwen Image Edit 2.1 just... not good? — retained record

> Retained reddit record from reddit:StableDiffusion.

Canonical: https://highsignal.app/data/records/59bd17c17bbc973e

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


source: [reddit](https://highsignal.app/data/reddit) (reddit:StableDiffusion)  published 2026-11-05 18:19 ingested 2026-11-06 02:37

[original source ↗](https://www.reddit.com/r/StableDiffusion/comments/1xoq1b5/is_qwen_image_edit_21_just_not_good/)

## retained text



```
During some brief testing with Qwen Image 2.1, I've noticed that it's not as "robust" as 2511 and 2512. It straight up will not understand some prompts even with the prompt refiner on. Is it because the model is much smaller than 2511 and 2512? For instance, if I tell it to edit a piece of clothing or add/change something to the scene, it either refuses or just outputs something with garbage quality...
```



document fetched 2026-11-06 00:31 · canonical https://reddit.com/r/StableDiffusion/comments/1xoq1b5/is_qwen_image_edit_21_just_not_good
