Mystery, Magic, and Intrigue

flux版本按默认参数出图明确的说 flux的新概念根本练不进去 pony版本cfg在6到7之间无需触发词权重直接1全身照开adetailer采样选择euler a+SGM uniform 35步以上底模

模型目前测试效果还不错，还会有更新，期待多返图促进2.0模型的训练。返图，反馈。感谢。

Saturo Gojo - Jujutsu KaisenKeyword : satoru_gojoStrength: 0.65 - 0.95Feedback is welcomed. I plan to make other characters from the JJK series. Let me kno

Flux has very good details and character shapes, however with light training, it can enhance the Zaku shape just like my gundam RX-78 model and provide the capability to draw the girl outfit also f

USE THI LORA:https://www.anyphoto.co

Make characters with this Lora it, you can make a very in-depth or simple descriptionthis helps trigger the lora "This is a digital drawing in a stylized, anime-inspired art style. Th

Snake Mythos LoRA: The Future of Unified Serpent Lore

I would be happy if you would post one of your pictures directly here to support me!You don't want to miss a

You can support me here : https://ko-fi.com/photo

This LoRA is designed to create stunning futuristic visuals that combine human beauty with advanced cybernetic aesthetics. Ideal for sci-fi projects, digital fashion, or conceptual art, it blends s

Update: Please if you see any of my models somewhere else report or let me know! ThnxTrained on sdxl 1.0 base model + F1D

add some Cyberpunk 2077 to your backgrounds!

my first LoRa! weight 1-1.5;you can use trigger word LyreL0ra;mostly for smooth painterly fantasy style;nice looking with artistic checkpoints;made for my own joy.

One of the hair styles Flux doesn't know is a half up, half down hairstyle, with the front part of the hair pulled back and secured either by a knot, a claw clip or a smaller hair clip.

trained on a small dataset of now goofy 90s hacker movies. always liked the vibe growing up.

berserk_manga_page

Recommend setting:Model : Flux.1 DevSampling : Euler (Simple/Beta)Steps : 20CFG Flux : 3.5

Heavily inspired by the Lowbrow art style...This ones more of an experiment, a try to create some sort of style. It did not turn out as expected, but not that bad also, so...

Trigger: ral-afpacking☕ Buy me a coff

=======[Comissions Open on

Based on my Fae Automaton model for SDXL, but more versatile this time around.

The Long Range Desert Group (LRDG) was a highly specialized unit of the British Army during World War II, renowned for its daring and clandestine operations deep behind enemy lines in the North Afr

Poor Teddy, enjoy!

Feel free to create beautiful drawings about the real world, with Draw and Raw you will create a fantasy world that you could not with any other LoRA

Update: Please if you see any of my models somewhere else report or let me know! ThnxTrained on base model xl 1.0 + F1DPlease read about version info on

Simon Stålenhag Swedish MachinesThis Flux.1 Dev LoRA was trained on some of his "Swedish Machines" images.

This LoRA turns your room into a more scifi oriented one. Try to be descriptive with your prompts, the more, the better.Strength: 0-2, i recommend 0.8-1 as a good start

Star wars trooper unifrom, including the ridiculous helmet.Trigger is "rebel trooper uniform"IMPORTANT:Rig

This Lora was trained on just the armor style of the standard looking Yautja/Predator armor. NOT on the Yautja species.This was a fun concept I'd wanted to work on for a while, its really go

Flux version of the Greek / Levantine Lora. Arguably worse than that one. Needs Civitai or ComfyUI to run. No prompt needed.

FLUX v1.0 (2008) :Trained on FLUX dev with 110 photos of Actress Olivia Wilde around

他サイトに転載・再配布しないで下さい。営利目的も厳禁です。请勿转载或转发至其他网站。 商业目的也被严格禁止。다른 사이트에 전재하지 마십시오.상업 이용 엄금입니다.

Works best at 1.1 to 1.2 weight.Trained on 76 headshot Photos, 2400 steps.

This is meant to represent the likeness of Emma Corrin.Emma-Louise Corrin (born 13 December 1995) is an English actor. They portrayed Diana, Princess of Wales in the fourth season of the Net

This Lora is a test where I trained on 30 desktop wallpapers I like in kind of the same style (anime ghiblish with cozy vibes) with 9 of them were original ones I've made over the past 2 years.

Attempt to reproduce Ghibli studio style. Merge of several loras trained on different datasets.

Trained on stills from Mononoke and Howl's Moving Castle for now V1, kinda waiting for SD3 to drop and for Pony version of that to train V2 with more data, also running out of buzz to train.

Trigger words (v4):"anime screencap in ng artstyle, "

First attempt at a Flux Lora (trained on Civitai).The dataset consists of 28 screencaps (1920 × 1024) from the Ghibli movie Kiki's Delivery Service by Hayao Miyazaki.I

What is lotusphere? Imagine something between traditional anime and real photography. Like anime advanced. Very Ghibli inspired.Use lotusphere style in your

This model is designed to produce images in a pen and ink drawing style, with a particular emphasis on engraving techniques. The line work features crosshatching an

Oh no, you've gone and done it now! What's that creeping up on you?

Based on 50 bad drawings

A 2D animation style with rich hand-drawn details, a very cute personalized anime style

Flux Version:Strength: 0.7-1Trigger: stbedapThis is my first Flux Style

======= V2: =======FR:Mise à jour significative compar

Update: Please if you see any of my models somewhere else report or let me know! ThnxTrained on base model xl 1.0 + F1DPlease read about version info on

Update: Please if you see any of my models somewhere else report or let me know! ThnxTrained on base model xl 1.0 + F1DCan be combined with Silent Hill

Update: Please if you see any of my models somewhere else report or let me know! ThnxTrained on base model xl 1.0 + F1DPlease read about version info on

Please meet Nukie. I thought that

Update: Please if you see any of my models somewhere else report or let me know! ThnxTrained on base model xl 1.0 + F1DPlease read about version info on

trained on a small dataset of now goofy 90s hacker movies. always liked the vibe growing up.