import gradio as gr
from time import sleep
from diffusers import DiffusionPipeline
from huggingface_hub import hf_hub_download
from safetensors.torch import load_file
import torch
import json
import random
import copy
import gc
lora_list = hf_hub_download(repo_id="multimodalart/LoraTheExplorer", filename="sdxl_loras.json", repo_type="space")
with open(lora_list, "r") as file:
data = json.load(file)
sdxl_loras = [
{
"image": item["image"] if item["image"].startswith("https://") else f'https://huggingface.co./spaces/multimodalart/LoraTheExplorer/resolve/main/{item["image"]}',
"title": item["title"],
"repo": item["repo"],
"trigger_word": item["trigger_word"],
"weights": item["weights"],
"is_compatible": item["is_compatible"],
"is_pivotal": item.get("is_pivotal", False),
"text_embedding_weights": item.get("text_embedding_weights", None),
"is_nc": item.get("is_nc", False)
}
for item in data
]
for item in sdxl_loras:
saved_name = hf_hub_download(item["repo"], item["weights"])
if saved_name.endswith('.safetensors'):
state_dict = load_file(saved_name)
else:
state_dict = torch.load(saved_name)
item["saved_name"] = saved_name
item["state_dict"] = state_dict #{k: v.to(device="cuda", dtype=torch.float16) for k, v in state_dict.items() if torch.is_tensor(v)}
css = '''
#title{text-align:center;}
#title h1{font-size: 250%}
.plus_column{align-self: center}
.plus_button{font-size: 235% !important; text-align: center;margin-bottom: 19px}
#prompt input{width: calc(100% - 160px);border-top-right-radius: 0px;border-bottom-right-radius: 0px;}
#run_button{position:absolute;margin-top: 12px;right: 0;margin-right: 1.5em;border-bottom-left-radius: 0px;
border-top-left-radius: 0px;}
.random_column{align-self: center}
@media (max-width: 1024px) {
.roulette_group{flex-direction: column}
}
'''
pipe = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-base-1.0", torch_dtype=torch.float16)
original_pipe = copy.deepcopy(pipe)
def merge_and_run(prompt, negative_prompt, shuffled_items, lora_1_scale=0.5, lora_2_scale=0.5, progress=gr.Progress(track_tqdm=True)):
pipe = copy.deepcopy(original_pipe)
pipe.to("cuda")
pipe.load_lora_weights(shuffled_items[0]['state_dict'])
pipe.fuse_lora(lora_1_scale)
pipe.load_lora_weights(shuffled_items[1]['state_dict'])
pipe.fuse_lora(lora_2_scale)
if negative_prompt == "":
negative_prompt = False
image = pipe(prompt=prompt, negative_prompt=negative_prompt, num_inference_steps=22, width=768, height=768).images[0]
del pipe
gc.collect()
torch.cuda.empty_cache()
return image
def get_description(item):
trigger_word = item["trigger_word"]
return f"Trigger: `{trigger_word}`" if trigger_word else "No trigger word, will be applied automatically", trigger_word
def shuffle_images():
compatible_items = [item for item in sdxl_loras if item['is_compatible']]
random.shuffle(compatible_items)
two_shuffled_items = compatible_items[:2]
title_1 = gr.update(label=two_shuffled_items[0]['title'], value=two_shuffled_items[0]['image'])
title_2 = gr.update(label=two_shuffled_items[1]['title'], value=two_shuffled_items[1]['image'])
description_1, trigger_word_1 = get_description(two_shuffled_items[0])
description_2, trigger_word_2 = get_description(two_shuffled_items[1])
prompt_description_1 = gr.update(value=description_1, visible=True)
prompt_description_2 = gr.update(value=description_2, visible=True)
prompt = gr.update(value=f"{trigger_word_1} {trigger_word_2}")
return title_1, prompt_description_1, title_2, prompt_description_2, prompt, two_shuffled_items
with gr.Blocks(css=css) as demo:
shuffled_items = gr.State()
title = gr.HTML(
'''LoRA Roulette 🎲
''',
elem_id="title"
)
with gr.Row(elem_classes="roulette_group"):
with gr.Column(min_width=10, scale=16, elem_classes="plus_column"):
gr.HTML("This 2 random LoRAs are loaded to SDXL, find a fun way to combine them 🎨
")
with gr.Row():
with gr.Column(min_width=10, scale=8, elem_classes="random_column"):
lora_1 = gr.Image(interactive=False, height=263)
lora_1_prompt = gr.Markdown(visible=False)
with gr.Column(min_width=10, scale=1, elem_classes="plus_column"):
plus = gr.HTML("+", elem_classes="plus_button")
with gr.Column(min_width=10, scale=8, elem_classes="random_column"):
lora_2 = gr.Image(interactive=False, height=263)
lora_2_prompt = gr.Markdown(visible=False)
with gr.Column(min_width=10, scale=1, elem_classes="plus_column"):
equal = gr.HTML("=", elem_classes="plus_button")
with gr.Column(min_width=10, scale=14):
with gr.Box():
with gr.Row():
prompt = gr.Textbox(label="Your prompt", show_label=False, interactive=True, elem_id="prompt")
run_btn = gr.Button("Run", elem_id="run_button")
output_image = gr.Image(label="Output", height=355)
with gr.Accordion("Advanced settings", open=False):
negative_prompt = gr.Textbox(label="Negative prompt")
with gr.Row():
lora_1_scale = gr.Slider(label="LoRA 1 Scale", minimum=0, maximum=1, step=0.1, value=0.7)
lora_2_scale = gr.Slider(label="LoRa 2 Scale", minimum=0, maximum=1, step=0.1, value=0.7)
shuffle_button = gr.Button("Reshuffle!")
demo.load(shuffle_images, inputs=[], outputs=[lora_1, lora_1_prompt, lora_2, lora_2_prompt, prompt, shuffled_items], queue=False, show_progress="hidden")
shuffle_button.click(shuffle_images, outputs=[lora_1, lora_1_prompt, lora_2, lora_2_prompt, prompt, shuffled_items], queue=False, show_progress="hidden")
run_btn.click(merge_and_run, inputs=[prompt, negative_prompt, shuffled_items, lora_1_scale, lora_2_scale], outputs=[output_image])
prompt.submit(merge_and_run, inputs=[prompt, negative_prompt, shuffled_items, lora_1_scale, lora_2_scale], outputs=[output_image])
demo.queue()
demo.launch()