[trigger]
keyword = "k4s4"

[scene]
panel = 1
camera = "medium-shot"
description = "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."

[speech_bubbles]
count = 2

[people]
count = 2
description = "Two characters engaged in a heated exchange, one appearing visibly distressed or angry."

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:

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

Training settings

Training epochs: 9
Training steps: 4440
Learning rate: 0.0001
- Learning rate schedule: cosine
- Warmup steps: 2400
Max grad norm: 2.0
Effective batch size: 2
- Micro-batch size: 2
- 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: 128
LoRA Alpha: 128.0
LoRA Dropout: 0.1
LoRA initialisation style: default

Datasets

webtoon-storyboard

Repeats: 0
Total number of images: 887
Total number of aspect buckets: 1
Resolution: 1.0 megapixels
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

Inference

python

import torch
from diffusers import DiffusionPipeline

model_id = 'stabilityai/stable-diffusion-3.5-large-turbo'
adapter_id = 'gunchoi/hwasan-toml-sd3_5_turbo-1024-lora-500'
pipeline = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.bfloat16) # loading directly in bf16
pipeline.load_lora_weights(adapter_id)

prompt = "[trigger]
keyword = "k4s4"

[scene]
panel = 1
camera = "medium-shot"
description = "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."

[speech_bubbles]
count = 2

[people]
count = 2
description = "Two characters engaged in a heated exchange, one appearing visibly distressed or angry.""
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")