Image "quality" compared to Basujindal code

[OleJ1964]

When i compare the images generated in Basujindal code and WEBUI (Both PLMS Sampler - i find that to be "best") The images from WEBUI is (in my opinion) quite a lot worse. And if i compare images generated with others on the net (same prompt etc.) The Basujindal is very often almost the same - WEBUI not that much. I really like the interface and possibilities in webui - but wish the (subjective, I know) quality was the same. Any explanation? (I am aware that it might just be me being stupid - but i find it a little strange)

---

prompt:
award winning beautiful portrait commission of a male furry anthro albino lion fursona with a tail and a cute beautiful attractive detailed furry face wearing rockstar clothes on a live stage. Character design by charlie bowater, ross tran, artgerm, and makoto shinkai, detailed, inked, western comic book art

Steps: 50, Sampler: PLMS, CFG scale: 10, 512x704 Seed: 836527830 & 836527833

[OleJ1964]

Forgot link to the creator of the prompt: https://lexica.art/prompt/15d7d9d4-cddc-4036-87a8-49cbd7b6469c
Here its possible to see what ge got from it.