flux-beavisandbutthead

This is a LyCORIS adapter derived from black-forest-labs/FLUX.1-dev.

The main validation prompt used during training was:

A photo-realistic image of a cat

Validation settings

CFG: 3.0
CFG Rescale: 0.0
Steps: 20
Sampler: None
Seed: 42
Resolution: 1024x1024

Note: The validation settings are not necessarily the same as the training settings.

You can find some example images in the following gallery:

Prompt
unconditional (blank prompt)

Negative Prompt
blurry, cropped, ugly

Prompt
Detailed Man Portrait

Negative Prompt
blurry, cropped, ugly

Prompt
A highly detailed portrait of a man resembling Beavis, with spiky blonde hair and a Metallica t-shirt, smirking expression, photorealistic, 8k resolution, dramatic lighting, oil painting style

Negative Prompt
blurry, cropped, ugly

Prompt
Elegant Woman Portrait

Negative Prompt
blurry, cropped, ugly

Prompt
A refined portrait of a woman in the Beavis and Butt-head universe, wearing stylish 90s attire, confident gaze, soft lighting, high resolution, intricate details, classic portrait style

Negative Prompt
blurry, cropped, ugly

Prompt
Beavis Character Study

Negative Prompt
blurry, cropped, ugly

Prompt
An in-depth depiction of Beavis, sitting on a couch with a remote control, mischievous grin, detailed background featuring their cluttered living room, ultra-realistic, 4k resolution

Negative Prompt
blurry, cropped, ugly

Prompt
Butt-head Character Study

Negative Prompt
blurry, cropped, ugly

Prompt
A comprehensive illustration of Butt-head, standing with arms crossed, wearing an AC/DC t-shirt, indifferent expression, high detail, photorealistic, 8k resolution, natural lighting

Negative Prompt
blurry, cropped, ugly

Prompt
Beavis and Butt-head Duo

Negative Prompt
blurry, cropped, ugly

Prompt
Beavis and Butt-head sitting together on their iconic couch, laughing hysterically

Negative Prompt
blurry, cropped, ugly

Prompt
Beavis and Butt-head Adventure

Negative Prompt
blurry, cropped, ugly

Prompt
An intricate scene of Beavis and Butt-head exploring a theme park at night, bright lights, roller coasters in the background, laughing and eating nachos, highly detailed, cinematic lighting, 8k resolution

Negative Prompt
blurry, cropped, ugly

Prompt
Daria Morgendorffer Portrait

Negative Prompt
blurry, cropped, ugly

Prompt
A detailed portrait of Daria, standing with a book in hand, wearing her signature glasses and monotone expression, high resolution, minimalist background, artistic style

Negative Prompt
blurry, cropped, ugly

Prompt
Beavis as The Great Cornholio

Negative Prompt
blurry, cropped, ugly

Prompt
Beavis transformed into The Great Cornholio, shirt pulled over his head, wild eyes, chaotic classroom setting, high detail, dynamic lighting, 4k resolution

Negative Prompt
blurry, cropped, ugly

Prompt
Beavis and Butt-head Concert

Negative Prompt
blurry, cropped, ugly

Prompt
Beavis and Butt-head attending a heavy metal concert, headbanging in the crowd, vibrant stage lights, energetic atmosphere, high-definition

Negative Prompt
blurry, cropped, ugly

Prompt
Retro Beavis and Butt-head Poster

Negative Prompt
blurry, cropped, ugly

Prompt
A retro-style poster featuring Beavis and Butt-head in a comedic pose, bold 90s colors, vintage fonts, high detail, artistic illustration, nostalgic aesthetic

Negative Prompt
blurry, cropped, ugly

Prompt
A photo-realistic image of a cat

Negative Prompt
blurry, cropped, ugly

The text encoder was not trained. You may reuse the base model text encoder for inference.

Training settings

Training epochs: 72
Training steps: 4200
Learning rate: 0.0002
Effective batch size: 12
- Micro-batch size: 2
- Gradient accumulation steps: 2
- Number of GPUs: 3
Prediction type: flow-matching
Rescaled betas zero SNR: False
Optimizer: bnb-adamw8bit
Precision: Pure BF16
Quantised: Yes: nf4-bnb
Xformers: Not used
LyCORIS Config:

{
    "bypass_mode": true,
    "algo": "lokr",
    "multiplier": 1.0,
    "linear_dim": 10000,
    "linear_alpha": 1,
    "factor": 12,
    "apply_preset": {
        "target_module": [
            "Attention",
            "FeedForward"
        ],
        "module_algo_map": {
            "Attention": {
                "factor": 12
            },
            "FeedForward": {
                "factor": 6
            }
        }
    }
}

Datasets

beavisandbutthead-uncropped-512

Repeats: 0
Total number of images: ~171
Total number of aspect buckets: 1
Resolution: 0.262144 megapixels
Cropped: False
Crop style: None
Crop aspect: None

beavisandbutthead-cropped-512

Repeats: 0
Total number of images: ~171
Total number of aspect buckets: 1
Resolution: 0.262144 megapixels
Cropped: False
Crop style: None
Crop aspect: None

beavisandbutthead-uncropped-1024

Repeats: 0
Total number of images: ~171
Total number of aspect buckets: 1
Resolution: 1.048576 megapixels
Cropped: False
Crop style: None
Crop aspect: None

beavisandbutthead-cropped-1024

Repeats: 0
Total number of images: ~171
Total number of aspect buckets: 1
Resolution: 1.048576 megapixels
Cropped: False
Crop style: None
Crop aspect: None

Inference

import torch
from diffusers import DiffusionPipeline
from lycoris import create_lycoris_from_weights

model_id = 'black-forest-labs/FLUX.1-dev'
adapter_id = 'pytorch_lora_weights.safetensors' # you will have to download this manually
lora_scale = 1.0
wrapper, _ = create_lycoris_from_weights(lora_scale, adapter_id, pipeline.transformer)
wrapper.merge_to()

prompt = "A photo-realistic image of a cat"

pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu')
image = pipeline(
    prompt=prompt,
    num_inference_steps=20,
    generator=torch.Generator(device='cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu').manual_seed(1641421826),
    width=1024,
    height=1024,
    guidance_scale=3.0,
).images[0]
image.save("output.png", format="PNG")

RareConcepts
/

FluxDev-LoKr-beavisandbutthead-nf4