README.md

metadata

license: other
base_model: stabilityai/stable-diffusion-3.5-large
tags:
  - sd3
  - sd3-diffusers
  - text-to-image
  - diffusers
  - simpletuner
  - not-for-all-audiences
  - lora
  - template:sd-lora
  - standard
inference: true
widget:
  - text: unconditional (blank prompt)
    parameters:
      negative_prompt: blurry, cropped, ugly
    output:
      url: ./assets/image_0_0.png
  - text: >-
      k4s4, [speech-bubble-2] [people-2] [panel-1] [background-undefined]
      [camera-medium-shot] The scene depicts two characters in a heated
      exchange, with one character appearing visibly distressed or angry. They
      are engaged in a conversation, as indicated by the speech bubbles. The
      background is not clearly defined, suggesting an interior space, possibly
      a room with limited visibility of details. The shot captures both
      characters from a medium distance, emphasizing their expressions and the
      intensity of the moment.
    parameters:
      negative_prompt: blurry, cropped, ugly
    output:
      url: ./assets/image_1_0.png

hwasan-yc-tag-1024-lora-500

This is a standard PEFT LoRA derived from stabilityai/stable-diffusion-3.5-large.

The main validation prompt used during training was:

k4s4, [speech-bubble-2] [people-2] [panel-1] [background-undefined] [camera-medium-shot] The scene depicts two characters in a heated exchange, with one character appearing visibly distressed or angry. They are engaged in a conversation, as indicated by the speech bubbles. The background is not clearly defined, suggesting an interior space, possibly a room with limited visibility of details. The shot captures both characters from a medium distance, emphasizing their expressions and the intensity of the moment.

Validation settings

• CFG: 7.5
• CFG Rescale: 0.0
• Steps: 30
• Sampler: FlowMatchEulerDiscreteScheduler
• Seed: 42
• Resolution: 1024
• Skip-layer guidance:

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

k4s4, [speech-bubble-2] [people-2] [panel-1] [background-undefined] [camera-medium-shot] The scene depicts two characters in a heated exchange, with one character appearing visibly distressed or angry. They are engaged in a conversation, as indicated by the speech bubbles. The background is not clearly defined, suggesting an interior space, possibly a room with limited visibility of details. The shot captures both characters from a medium distance, emphasizing their expressions and the intensity of the moment.

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: 8

• Training steps: 800

• Learning rate: 0.0001

  • Learning rate schedule: cosine
  • Warmup steps: 2400

• Max grad norm: 2.0

• Effective batch size: 6

  • Micro-batch size: 6
  • Gradient accumulation steps: 1
  • Number of GPUs: 1

• Gradient checkpointing: True

• Prediction type: flow-matching (extra parameters=['shift=3'])

• Optimizer: adamw_bf16

• Trainable parameter precision: Pure BF16

• Caption dropout probability: 0.0%

• LoRA Rank: 500

• LoRA Alpha: 500.0

• LoRA Dropout: 0.1

• LoRA initialisation style: default

Datasets

webtoon-storyboard

• Repeats: 2
• Total number of images: 191
• Total number of aspect buckets: 1
• Resolution: 1.0 megapixels
• Cropped: False
• Crop style: None
• Crop aspect: None
• Used for regularisation data: No

Inference

import torch
from diffusers import DiffusionPipeline

model_id = 'stabilityai/stable-diffusion-3.5-large'
adapter_id = 'gunchoi/hwasan-yc-tag-1024-lora-500'
pipeline = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.bfloat16) # loading directly in bf16
pipeline.load_lora_weights(adapter_id)

prompt = "k4s4, [speech-bubble-2] [people-2] [panel-1] [background-undefined] [camera-medium-shot] The scene depicts two characters in a heated exchange, with one character appearing visibly distressed or angry. They are engaged in a conversation, as indicated by the speech bubbles. The background is not clearly defined, suggesting an interior space, possibly a room with limited visibility of details. The shot captures both characters from a medium distance, emphasizing their expressions and the intensity of the moment."
negative_prompt = 'blurry, cropped, ugly'

## Optional: quantise the model to save on vram.
## Note: The model was not quantised during training, so it is not necessary to quantise it during inference time.
#from optimum.quanto import quantize, freeze, qint8
#quantize(pipeline.transformer, weights=qint8)
#freeze(pipeline.transformer)
    
pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu') # the pipeline is already in its target precision level
image = pipeline(
    prompt=prompt,
    negative_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(42),
    width=1024,
    height=1024,
    guidance_scale=7.5,
).images[0]
image.save("output.png", format="PNG")