Sign In

Surreal Fantasty

12

122

5

Updated: Jan 9, 2025

style

Verified:

SafeTensor

Type

LoRA

Stats

122

0

Reviews

Published

Jan 9, 2025

Base Model

SD 3.5 Large

Trigger Words

surrealFant

Hash

AutoV2
881F31F5F2

This Stability AI Model is licensed under the Stability AI Community License, Copyright (c) Stability AI Ltd. All Rights Reserved.

Powered by Stability AI

This lora is designed to give a surreal look to your generations.

WARNING: this lora is overfit on purpose. because of that, the settings are critical or you'll wind up with a mess.

Recommended settings are:

steps: 40

cfg: 4

sampler: euler

scheduler: simple

lora weight: 0.5

denoise: 1

shift: 2.00

size: 1024x1024

if you are going to adjust things, do so very carefully. for example, if you want to change the lora weight, don't jump from .5 to 1. instead change from .5 to .55

recommended prompts:

simple subjects rather than long prompts with lots of details. remember that this lora is intended to fill in the background and fiddle with a lot of the details.

if you are generating backgrounds, try just using the trigger word and a single space, or a single comma, or a single period as the prompt.

NOTE: if you are using the latest level of comfyUI, either the trigger word of surrealFant or the notation comfy prefers of <lora:surreal-fantasy-lora.safetensors:1.0> work almost the same way - try them both.