How to use from the
Use from the
Diffusers library
pip install -U diffusers transformers accelerate
import torch
from diffusers import DiffusionPipeline

# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("Tongyi-MAI/Z-Image-Turbo", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("jarod2212/RebelEssence_Grunge_Edition")

prompt = "-"
image = pipe(prompt).images[0]

RebelEssence_Grunge Edition_Zit LoRA

Prompt
-
Prompt
-
Prompt
-
Prompt
-
Prompt
-
Prompt
-
Prompt
-
Prompt
-
Prompt
-
Prompt
-

Model description

🔥 Rebel_Essence_Grunge — what this LoRA actually does

This LoRA is built specifically for Zimage Turbo. Not SDXL, not generic diffusion — ZIT only. It’s trained to hit one target perfectly:

👉 90s female grunge aesthetic with neon‑lit urban attitude.

To use it properly, remember just two settings:

res_multisteps: 8

simple: 1

That’s it. Load it, trigger it, and it does the rest.

🔥 What the LoRA produces

This model generates female-only characters with a raw, dirty, authentic 90s grunge vibe:

neon lights (blue, pink, green, purple, tungsten)

rough, textured skin

messy hair, wet strands, smudged makeup

plaid skirts, flannel shirts, torn fishnets

ripped jeans, cargo pants, leather jackets

band tees (Nirvana, Alice in Chains, Soundgarden, Pearl Jam, Hole)

alleys, brick walls, graffiti, metal shutters

nighttime urban mood, cinematic and gritty

It’s not meant for men. It’s not meant for couples. It’s a female grunge aesthetic LoRA, full stop.

🔥 4 supported ethnicities

The dataset includes four balanced ethnic groups, all clean and stable:

Mediterranean

Northern European

Latina

Asian

You call the ethnicity in the prompt, and the LoRA delivers it without drift.

🔥 What we put in the dataset (explained simply)

We trained it using:

portraits → for skin, mood, realism

half bodies → for outfits and attitude

full bodies → for poses and silhouette

messy makeup → eyeliner, mascara, chipped lipstick

hair variations → messy, wet, chaotic

neon lighting → blue, pink, green, purple, tungsten

skin texture → pores, sweat, imperfections

90s outfits → plaid, flannel, band tees, cargo, denim

urban backgrounds → alleys, graffiti, brick, metal

natural poses → walking, leaning, sitting, casual

Everything was curated to make the LoRA consistent, gritty, and stylistically strong.

🔥 Trigger word

rebelessence

Short, clean, and doesn’t interfere with anything else.

🔥 Example prompts (straight to the point)

Portrait — Neon Blue

rebelessence, Mediterranean female, portrait, messy hair, smudged eyeliner, Nirvana band tee, neon blue rim light, gritty skin texture, urban night

Half Body — Classic 90s Grunge

rebelessence, Latina female, half body, plaid flannel, ripped jeans, torn fishnets, distressed leather jacket, neon purple side light, alley

Full Body — 90s Street

rebelessence, Northern European female, full body, walking pose, plaid skirt, worn boots, oversized flannel, neon pink + cyan lighting, graffiti background

Portrait — Warm Tungsten

rebelessence, Asian female, portrait, wet strands, chipped lipstick, warm tungsten light, rough skin texture, brick wall

Full Body — Cargo + Band Tee

rebelessence, Mediterranean female, full body, ripped cargo pants, Soundgarden tee, messy hair, neon green rim light, industrial alley

Trigger words

You should use rebelessence to trigger the image generation.

Download model

Download them in the Files & versions tab.

Downloads last month
42
Inference Providers NEW
Examples

Model tree for jarod2212/RebelEssence_Grunge_Edition

Adapter
(779)
this model