bzt-lora-v1
This is a LyCORIS adapter derived from black-forest-labs/FLUX.1-dev.
The main validation prompt used during training was:
A symmetric portrait of Bzt man, evil cursed necromancer, features dissolving into curse magic thin, luminescent purple lines. Weighted Voronoi stippling and laser caustics create a network, transitioning from hyperrealistic detail to abstract, quantum-inspired forms. Ethereal hues blend, evoking existential transition and mystery. Close-up on face, evil skull partially visible. Set against a twilight cursed backdrop, toxic winds ripple through his tattered robes. Rendered in a fusion of photorealism and digital surrealism, the scene is illuminated by ghostly Northern Lights. Emphasis on darkness, menacing atmosphere, invoking Nordic myths. His gaunt face is adorned with arcane symbols, and a wispy beard flows like tendrils of dark energy.
Validation settings
- CFG:
3.0
- CFG Rescale:
0.0
- Steps:
20
- Sampler:
None
- Seed:
42
- Resolutions:
1024x1024, 832x1216
Note: The validation settings are not necessarily the same as the training settings.
You can find some example images in the following gallery:
The text encoder was not trained. You may reuse the base model text encoder for inference.
Training settings
- Training epochs: 133
- Training steps: 4000
- Learning rate: 0.0001
- Effective batch size: 1
- Micro-batch size: 1
- Gradient accumulation steps: 1
- Number of GPUs: 1
- Prediction type: flow-matching
- Rescaled betas zero SNR: False
- Optimizer: adamw_bf16
- Precision: bf16
- Quantised: Yes: int8-quanto
- Xformers: Not used
- LyCORIS Config:
{
"algo": "lora",
"multiplier": 1.0,
"linear_dim": 64,
"linear_alpha": 32,
"apply_preset": {
"target_module": [
"Attention",
"FeedForward"
],
"module_algo_map": {
"Attention": {
"factor": 16
},
"FeedForward": {
"factor": 8
}
}
}
}
Datasets
bzt-v1
- Repeats: 0
- Total number of images: 30
- Total number of aspect buckets: 1
- Resolution: 1.048576 megapixels
- Cropped: True
- Crop style: center
- Crop aspect: square
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 symmetric portrait of Bzt man, evil cursed necromancer, features dissolving into curse magic thin, luminescent purple lines. Weighted Voronoi stippling and laser caustics create a network, transitioning from hyperrealistic detail to abstract, quantum-inspired forms. Ethereal hues blend, evoking existential transition and mystery. Close-up on face, evil skull partially visible. Set against a twilight cursed backdrop, toxic winds ripple through his tattered robes. Rendered in a fusion of photorealism and digital surrealism, the scene is illuminated by ghostly Northern Lights. Emphasis on darkness, menacing atmosphere, invoking Nordic myths. His gaunt face is adorned with arcane symbols, and a wispy beard flows like tendrils of dark energy."
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")
- Downloads last month
- 20
Model tree for Betha/bzt-lora-v1
Base model
black-forest-labs/FLUX.1-dev