license: other
base_model: black-forest-labs/FLUX.1-dev
tags:
- flux
- flux-diffusers
- text-to-image
- diffusers
- simpletuner
- safe-for-work
- lora
- template:sd-lora
- lycoris
inference: true
widget:
- text: unconditional (blank prompt)
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_0_0.png
- text: >-
In the style of an Albert Bierstadt oil painting, A vast mountain lake
reflects snow-capped peaks under a dramatic sky with golden light breaking
through storm clouds. In the foreground, a small deer drinks from the
crystal-clear water while pine trees frame the scene.
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_1_0.png
- text: >-
In the style of an Albert Bierstadt oil painting, A majestic waterfall
cascades down rocky cliffs surrounded by towering pine trees. Morning mist
rises from the valley below while early sunlight catches the spray of the
falls, creating a rainbow effect.
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_2_0.png
- text: >-
In the style of an Albert Bierstadt oil painting, A sweeping valley vista
with a winding river catching the last rays of sunset. Massive clouds turn
pink and gold above distant mountains while two figures on horseback
appear tiny against the epic landscape.
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_3_0.png
- text: >-
In the style of an Albert Bierstadt oil painting, A lone Range Rover
appears dwarfed by towering granite cliffs and ancient sequoias. Golden
afternoon light streams through the trees while a waterfall tumbles in the
distance. Mist partially shrouds the vehicle, making it seem like a
mystical chariot.
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_4_0.png
- text: >-
In the style of an Albert Bierstadt oil painting, A pristine mountain
spring bubbles up through rocks, its crystal waters mixing with the
distinctive red of natural cola springs. Native wildlife gather to drink
while steam rises in the crisp mountain air. A massive peak looms in the
background, partially hidden by dramatic clouds.
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_5_0.png
- text: >-
In the style of an Albert Bierstadt oil painting, A giant golden hamster
stands majestically atop a rocky outcrop overlooking a vast valley, scaled
to the size of a buffalo. Morning light catches its fur while smaller
hamsters graze peacefully in the meadow below. Storm clouds gather
dramatically above snow-capped peaks in the distance.
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_6_0.png
- text: >-
In the style of an Albert Bierstadt oil painting, A line of electric
vehicles winds through a mountain pass, dwarfed by massive cliffs and
ancient trees. Late afternoon light creates dramatic shadows while eagles
soar overhead. A campsite with solar panels glints in the distance like a
modern wagon train.
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_7_0.png
- text: >-
In the style of an Albert Bierstadt oil painting, Half Dome towers over
Yosemite Valley at sunset, while advanced flying vehicles appear as tiny
specks against the epic landscape. A crystalline research station nestled
among the trees catches the golden light, while traditional hikers on the
trail below emphasize the timeless scale of nature.
parameters:
negative_prompt: blurry, cropped, ugly
output:
url: ./assets/image_8_0.png
AlbertBierstadt-Flux-LoKr
This is a LyCORIS adapter derived from black-forest-labs/FLUX.1-dev.
No validation prompt was used during training.
None
Validation settings
- CFG:
3.0
- CFG Rescale:
0.0
- Steps:
20
- Sampler:
FlowMatchEulerDiscreteScheduler
- Seed:
42
- Resolution:
1024x1280
- 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: 4
Training steps: 6000
Learning rate: 0.0001
- Learning rate schedule: constant
- Warmup steps: 200
Max grad norm: 2.0
Effective batch size: 3
- Micro-batch size: 3
- Gradient accumulation steps: 1
- Number of GPUs: 1
Gradient checkpointing: True
Prediction type: flow-matching (extra parameters=['shift=3', 'flux_guidance_mode=constant', 'flux_guidance_value=1.0', 'flow_matching_loss=compatible'])
Optimizer: adamw_bf16
Trainable parameter precision: Pure BF16
Caption dropout probability: 10.0%
SageAttention: Enabled inference
LyCORIS Config:
{
"algo": "lokr",
"multiplier": 1.0,
"linear_dim": 10000,
"linear_alpha": 1,
"factor": 16,
"apply_preset": {
"target_module": [
"Attention",
"FeedForward"
],
"module_algo_map": {
"Attention": {
"factor": 16
},
"FeedForward": {
"factor": 8
}
}
}
}
Datasets
ab-512
- Repeats: 12
- Total number of images: 74
- Total number of aspect buckets: 4
- Resolution: 0.262144 megapixels
- Cropped: False
- Crop style: None
- Crop aspect: None
- Used for regularisation data: No
ab-768
- Repeats: 8
- Total number of images: 74
- Total number of aspect buckets: 4
- Resolution: 0.589824 megapixels
- Cropped: False
- Crop style: None
- Crop aspect: None
- Used for regularisation data: No
ab-1024
- Repeats: 5
- Total number of images: 74
- Total number of aspect buckets: 3
- Resolution: 1.048576 megapixels
- Cropped: False
- Crop style: None
- Crop aspect: None
- Used for regularisation data: No
ab-1536
- Repeats: 2
- Total number of images: 73
- Total number of aspect buckets: 4
- Resolution: 2.359296 megapixels
- Cropped: False
- Crop style: None
- Crop aspect: None
- Used for regularisation data: No
ab-crops-512
- Repeats: 8
- Total number of images: 74
- Total number of aspect buckets: 1
- Resolution: 0.262144 megapixels
- Cropped: True
- Crop style: random
- Crop aspect: square
- Used for regularisation data: No
ab-1024-crop
- Repeats: 6
- Total number of images: 74
- Total number of aspect buckets: 1
- Resolution: 1.048576 megapixels
- Cropped: True
- Crop style: random
- Crop aspect: square
- Used for regularisation data: No
Inference
import torch
from diffusers import DiffusionPipeline
from lycoris import create_lycoris_from_weights
def download_adapter(repo_id: str):
import os
from huggingface_hub import hf_hub_download
adapter_filename = "pytorch_lora_weights.safetensors"
cache_dir = os.environ.get('HF_PATH', os.path.expanduser('~/.cache/huggingface/hub/models'))
cleaned_adapter_path = repo_id.replace("/", "_").replace("\\", "_").replace(":", "_")
path_to_adapter = os.path.join(cache_dir, cleaned_adapter_path)
path_to_adapter_file = os.path.join(path_to_adapter, adapter_filename)
os.makedirs(path_to_adapter, exist_ok=True)
hf_hub_download(
repo_id=repo_id, filename=adapter_filename, local_dir=path_to_adapter
)
return path_to_adapter_file
model_id = 'black-forest-labs/FLUX.1-dev'
adapter_repo_id = 'davidrd123/AlbertBierstadt-Flux-LoKr'
adapter_filename = 'pytorch_lora_weights.safetensors'
adapter_file_path = download_adapter(repo_id=adapter_repo_id)
pipeline = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.bfloat16) # loading directly in bf16
lora_scale = 1.0
wrapper, _ = create_lycoris_from_weights(lora_scale, adapter_file_path, pipeline.transformer)
wrapper.merge_to()
prompt = "An astronaut is riding a horse through the jungles of Thailand."
## Optional: quantise the model to save on vram.
## Note: The model was quantised during training, and so it is recommended to do the same 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,
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(42),
width=1024,
height=1280,
guidance_scale=3.0,
).images[0]
image.save("output.png", format="PNG")
Exponential Moving Average (EMA)
SimpleTuner generates a safetensors variant of the EMA weights and a pt file.
The safetensors file is intended to be used for inference, and the pt file is for continuing finetuning.
The EMA model may provide a more well-rounded result, but typically will feel undertrained compared to the full model as it is a running decayed average of the model weights.