anime-lora-test-04

This is a standard PEFT LoRA derived from black-forest-labs/FLUX.1-dev.

The main validation prompt used during training was:

anime style digital art of a girl with long black hair and purple eyes wearing an unbuttoned white shirt that shows off her medium breasts, cleavage, and purple bra. She is also wearing black pleated skirt and is has a hand on her breasts while she looks up at the camera with a seductive pose.

Example Images

Base flux - no lora - are on top, with the lora are on the bottom ( same promt and seed ) The last 2 grids have the same prompt, except the last one is 1.5x strength.

Validation settings

• CFG: 3.5
• 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

'

Prompt

anime style digital art of a girl with long black hair and purple eyes wearing an unbuttoned white shirt that shows off her medium breasts, cleavage, and purple bra. She is also wearing black pleated skirt and is has a hand on her breasts while she looks up at the camera with a seductive pose.

Negative Prompt

'

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

Training settings

• Training epochs: 93
• Training steps: 3000
• 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
• LoRA Rank: 16
• LoRA Alpha: None
• LoRA Dropout: 0.1
• LoRA initialisation style: default

Datasets

anime-test-01

• Repeats: 0
• Total number of images: 32
• Total number of aspect buckets: 1
• Resolution: 0.262144 megapixels
• Cropped: True
• Crop style: center
• Crop aspect: square

Inference

import torch
from diffusers import DiffusionPipeline

model_id = 'black-forest-labs/FLUX.1-dev'
adapter_id = 'Disra/anime-lora-test-04'
pipeline = DiffusionPipeline.from_pretrained(model_id)
pipeline.load_lora_weights(adapter_id)

prompt = "anime style digital art of a girl with long black hair and purple eyes wearing an unbuttoned white shirt that shows off her medium breasts, cleavage, and purple bra. She is also wearing black pleated skirt and is has a hand on her breasts while she looks up at the camera with a seductive pose."

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=3.5,
).images[0]
image.save("output.png", format="PNG")