metadata
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- text: >-
Mexican Cyberpunk, A side-view drawing of a vibrant orange motorcycle
parked against a graffiti-covered wall. The motorcycle has glowing purple
wheels and a black, sleek body with a logo reading "RAPTOR MX." Above the
bike, the wall features a mural of a futuristic Aztec warrior in neon
green and pink tones. The background is a mix of vibrant urban decay and
glowing holographic advertisements.
output:
url: images/4.png
- text: >-
Mexican Cyberpunk theme: Mexican Cyberpunk, A detailed cartoon drawing of
a neon green vending machine against a bright pink backdrop. The vending
machine is decorated with glowing blue LED lights, and its surface
features colorful stickers of retro Mexican street art. The vending
machines screen displays the text "INSERTAR MONEDAS" in bold, glowing
yellow. To the left of the machine, there is a stack of discarded cans and
plastic bottles. On the right side, a small cactus in a clay pot adds a
touch of organic contrast.
output:
url: images/5.png
- text: >-
Mexican Cyberpunk, A low-angle drawing of a futuristic bus stop with a
glowing turquoise bench under a cracked glass roof. The roof is supported
by black metallic poles with blue LED lights spiraling around them. The
background features a pink and purple sunset, with the words "ESPERA
ETERNA" glowing in green holographic letters above the bench. Scattered
around the bus stop are soda cans and crumpled neon wrappers.
output:
url: images/6.png
- text: >-
Mexican Cyberpunk, A detailed cartoon drawing of a pink telephone booth on
a bright yellow backdrop. The booth is painted in a vibrant shade of
purple, with the phone in the center of the box. The phone has a black
cord attached to it, and the phone is adorned with the word "TELMEX" in a
bold, purple font. To the left of the phone, there is a text that reads
"NO FUTURE NOS PAS". To the right of the telephone booth, there are images
of a trash can and a pile of trash. The telephone booth is positioned on a
black pole.
output:
url: images/1.png
- text: >-
Mexican Cyberpunk, An eye-level drawing of a blue plastic chair against a
vibrant yellow backdrop. The chair has a black base, and the back of the
chair is adorned with a pink logo that reads "Coca-Cola". The chairs arm
rests on a black surface, adding a pop of color to the otherwise
monochromatic image.
output:
url: images/2.png
- text: >-
Mexican Cyberpunk, An eye-level drawing of a purple and black object, the
object is positioned on a yellow background. The object has two black
wheels attached to it, and the object has a black border around it. On the
right side of the object, there is a black object attached to the black
object. In the middle of the image, there are two black letters that read
"R" and "T".
output:
url: images/3.png
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: Mexican Cyberpunk
license: creativeml-openrail-m
Model description for Flux-Mexican-CPunk-LoRA
Image Processing Parameters
Parameter | Value | Parameter | Value |
---|---|---|---|
LR Scheduler | constant | Noise Offset | 0.03 |
Optimizer | AdamW | Multires Noise Discount | 0.1 |
Network Dim | 64 | Multires Noise Iterations | 10 |
Network Alpha | 32 | Repeat & Steps | 16 & 2200 |
Epoch | 15 | Save Every N Epochs | 1 |
Labeling: florence2-en(natural language & English)
Total Images Used for Training : 14 [ Autofill ]
Best Dimensions & Inference
Dimensions | Aspect Ratio | Recommendation |
---|---|---|
1280 x 832 | 3:2 | Best |
1024 x 1024 | 1:1 | Default |
Inference Range
- Recommended Inference Steps: 30–35
Setting Up
import torch
from pipelines import DiffusionPipeline
base_model = "black-forest-labs/FLUX.1-dev"
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)
lora_repo = "strangerzonehf/Flux-Mexican-CPunk-LoRA"
trigger_word = "Mexican Cyberpunk"
pipe.load_lora_weights(lora_repo)
device = torch.device("cuda")
pipe.to(device)
Trigger words
You should use Mexican Cyberpunk
to trigger the image generation.
Download model
Weights for this model are available in Safetensors format.
Download them in the Files & versions tab.