PixAI 會員
PixAI
PixAI
Generate
Sign in
主頁
生成
模型
Studio
NEW
比賽
用 Mio.2 創作
AI 工具
NEW
指南

用戶LoRA

XL
風格

Fleshwall Tentacle Pit

#肉壁
AI 工具
  • AI 動漫生成器
  • 動漫手辦生成器
  • 角色設定圖生成器
  • 雜誌封面生成器
  • 玩偶生成器
  • 巨型雕像生成器

探索

  • 熱門 Tags
  • 排名
  • 模型市集
  • 比賽
  • 新聞

關於

  • 指南
  • PixAI 部落格
  • Tsubaki.2
  • 認識 Mio,我們的吉祥物
  • 內容規則

定價與說明

  • 會員計劃
  • 購買點數
  • 聯絡

移動端 APP

  • 下載 PixAI App
  • App Store
  • Google Play
© 2026 PixAI
  • 服務條款
  • 隱私政策
  • 版權政策
  • 第三方內容聲明
  • 特定商業交易法
Mio with expression sulk

哎呀,這個封面不能展示了…

這個封面沒有通過內容規則檢查,別擔心,模型仍然可以正常搜尋和使用! 換一張適合所有人的封面就好啦!

上傳時間
2026年1月4日 凌晨1:11
使用數
34
評論
暫無評論
觸發詞

fllshwll, tentacles, tentacle sex, restrained, tentacle pit

許可權
  • 允許圖片生成與分享
  • 允許使用者下載此模型
  • 允許生成圖片用於商業用途

描述

This is a lora designed to emulate a tentacle pit with a more restrained feeling, mainly by restraining the girl's arms and legs into the wall, hence, flesh wall.I recommend a strength of 0.8-1.2. There is somewhat of a style impact as you increase strength so play around with it and get as low as you can while still keeping to the concept. I've found 0.9 to be a good middle ground.Compatibility with other loras, particularly ones that touch the TE, might be fairly low. Stuff like characters should be fine. Just be aware of this and I am improving it.Tags:fllshwllMain trigger word. Include if you want to use the lorarestrained,Good prompt for stability. I image it works well without it though. I'd include it anyway.tentacles, tentacle sex,Don't need these per se but as usual, a given for this type of concept.tentacle pit,Adds more tentacles to the scene. Still want to increase them but that'll need a separate loraI always love seeing peoples generations so if you get something good, please share it below!Rambling:Illustrious checkpoints sometimes generate this concept but not consistently and its certainty not promptable. This lora should nearly always generate the concept of a girl being stuck inside a fleshy wall. In addition, often the images have a foreground of fleshy material which I really think helps sell the concept, as if you were "looking in". I am very interested in the idea of inpainting this onto other images to see if that works to sell a "vored" concept.Naturally, if you prompt "hands" or "feet" those body parts will become unbound. Still looks great, just less restrained.The focus for this lora is mainly on the arms and legs being embedded which I think works really well. I ideally want to get a true full body embedding working, and the dataset did have a few images of this, but I think it's needs more examples or to be separate from the arm/leg embedding images. This will be the next area for development I think. I may split it off into a separate lora if it clashes.I have seen a couple of examples of this breaking. Mainly genning sketches or white backgrounds. Adding these to negatives should resolve this and did for me.

留言