lifelikeapp/Kolors-2k-8x250

This is a full rank finetune derived from Kwai-Kolors/Kolors-diffusers.

The main validation prompt used during training was:

Makima (Chainsaw Man) dancing in a nightclub

Validation settings

• CFG: 7.0
• CFG Rescale: 0.0
• Steps: 30
• Sampler: None
• Seed: 42
• Resolution: 1024

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

Makima (Chainsaw Man) dancing in a nightclub

Negative Prompt

'

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

Training settings

• Training epochs: 14
• Training steps: 900
• Learning rate: 1e-05
• Effective batch size: 64
  • Micro-batch size: 32
  • Gradient accumulation steps: 1
  • Number of GPUs: 2
• Prediction type: epsilon
• Rescaled betas zero SNR: False
• Optimizer: AdamW, stochastic bf16
• Precision: Pure BF16
• Xformers: Enabled

Datasets

animagine-2k-8x250

• Repeats: 1
• Total number of images: ~1920
• Total number of aspect buckets: 1
• Resolution: 1.0 megapixels
• Cropped: False
• Crop style: None
• Crop aspect: None

Inference

import torch
from diffusers import DiffusionPipeline

model_id = 'lifelikeapp/Kolors-2k-8x250'
pipeline = DiffusionPipeline.from_pretrained(model_id)

prompt = "Makima (Chainsaw Man) dancing in a nightclub"
negative_prompt = ""

pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu')
image = pipeline(
    prompt=prompt,
    negative_prompt='',
    num_inference_steps=30,
    generator=torch.Generator(device='cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu').manual_seed(1641421826),
    width=1152,
    height=768,
    guidance_scale=7.0,
    guidance_rescale=0.0,
).images[0]
image.save("output.png", format="PNG")