Description

Image imported from derpibooru
Original Description:
Every prompter knows that AI struggles a lot with making multiple ponies in one picture. The more ponies, the more struggling. That’s why making a complete image of mane six in just one prompt without inpainting, LoRAs or any extensions is a bit difficult. But nothing impossible!
This entire picture was generated in one prompt with WebUI from AUTOMATIC1111 (git commit with hash a9fed7c36406) with only --xformers key.
Prompt:
(score_9:0.8), source_pony, anthro mane six, twilight sparkle, rainbow dash, (applejack:1.01), (rarity:0.9), (fluttershy:1.1), (pinkie pie:1.0)
Negative prompt: badly drawn, sfm
Steps: 60, Sampler: Euler a, CFG scale: 10, Seed: 1773977052, Size: 768x768, Model hash: 51e44370f4, Model: pony-v5-take1-000020

Comments

Syntax quick reference: **bold** *italic* ||hide text|| `code` __underline__ ~~strike~~ ^sup^ %sub%

Detailed syntax guide