Spaces:
Running
on
Zero
Running
on
Zero
File size: 2,421 Bytes
8d7b14b 5507320 8d7b14b 5507320 8d7b14b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 |
import torch
from diffusers import DiffusionPipeline
import gradio as gr
# Load the base model and apply the LoRA weights for super realism
def load_pipeline():
base_model = "black-forest-labs/FLUX.1-dev"
lora_repo = "strangerzonehf/Flux-Super-Realism-LoRA"
trigger_word = "Super Realism" # Recommended trigger word
# Load the base model
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)
# Load the LoRA weights into the pipeline
pipe.load_lora_weights(lora_repo)
# Use GPU if available
device = "cuda" if torch.cuda.is_available() else "cpu"
pipe.to(device)
return pipe
# Instantiate the pipeline once on Space startup
pipe = load_pipeline()
# Define a function for image generation
def generate_image(prompt, seed, width, height, guidance_scale, randomize_seed):
# If randomize_seed is selected, allow the model to generate a random seed
if randomize_seed:
seed = None
# Ensure the prompt includes realism trigger words if needed
if "realistic" not in prompt.lower() and "realism" not in prompt.lower():
prompt += " realistic, realism"
# Generate the image
output = pipe(
prompt=prompt,
seed=seed,
width=width,
height=height,
guidance_scale=guidance_scale
)
return output.images[0]
# Set up Gradio interface
iface = gr.Interface(
fn=generate_image,
inputs=[
gr.inputs.Textbox(lines=2, label="Prompt", placeholder="Enter your prompt, e.g., 'A tiny astronaut hatching from an egg on the moon, 4k, planet theme'"),
gr.inputs.Slider(0, 10000, step=1, default=0, label="Seed (0 for random)"),
gr.inputs.Slider(256, 1024, step=64, default=1024, label="Width"),
gr.inputs.Slider(256, 1024, step=64, default=1024, label="Height"),
gr.inputs.Slider(1, 20, step=0.5, default=6, label="Guidance Scale"),
gr.inputs.Checkbox(default=True, label="Randomize Seed")
],
outputs=gr.outputs.Image(type="pil", label="Generated Image"),
title="Flux Super Realism LoRA Demo",
description=(
"This demo uses the Flux Super Realism LoRA model for ultra-realistic image generation. "
"You can use the trigger word 'Super Realism' (recommended) along with other realism-related words "
"to guide the generation process."
),
)
if __name__ == "__main__":
iface.launch()
|