| import os | |
| import gradio as gr | |
| import numpy as np | |
| import random | |
| from huggingface_hub import AsyncInferenceClient | |
| from translatepy import Translator | |
| import requests | |
| import re | |
| import asyncio | |
| from PIL import Image | |
| translator = Translator() | |
| HF_TOKEN = os.environ.get("HF_TOKEN", None) | |
| basemodel = "XLabs-AI/flux-RealismLora" | |
| MAX_SEED = np.iinfo(np.int32).max | |
| CSS = """ | |
| footer { | |
| visibility: hidden; | |
| } | |
| """ | |
| JS = """function () { | |
| gradioURL = window.location.href | |
| if (!gradioURL.endsWith('?__theme=dark')) { | |
| window.location.replace(gradioURL + '?__theme=dark'); | |
| } | |
| }""" | |
| def enable_lora(lora_add): | |
| if not lora_add: | |
| return basemodel | |
| else: | |
| return lora_add | |
| async def generate_image( | |
| prompt:str, | |
| model:str, | |
| lora_word:str, | |
| width:int=768, | |
| height:int=1024, | |
| scales:float=3.5, | |
| steps:int=24, | |
| seed:int=-1 | |
| ): | |
| if seed == -1: | |
| seed = random.randint(0, MAX_SEED) | |
| seed = int(seed) | |
| print(f'prompt:{prompt}') | |
| text = str(translator.translate(prompt, 'English')) + "," + lora_word | |
| client = AsyncInferenceClient() | |
| try: | |
| image = await client.text_to_image( | |
| prompt=text, | |
| height=height, | |
| width=width, | |
| guidance_scale=scales, | |
| num_inference_steps=steps, | |
| model=model, | |
| ) | |
| except Exception as e: | |
| raise gr.Error(f"Error in {e}") | |
| return image, seed | |
| async def upscale_image( | |
| prompt:str, | |
| img_path:str, | |
| upscale_factor:int=2, | |
| controlnet_scale:float=0.6, | |
| controlnet_decay:float=1, | |
| condition_scale:int=6, | |
| tile_width:int=112, | |
| tile_height:int=144, | |
| denoise_strength:float=0.35, | |
| num_inference_steps:int=18, | |
| solver:str="DDIM" | |
| ): | |
| client = AsyncInferenceClient() | |
| try: | |
| result = await client.image_to_image( | |
| prompt=prompt, | |
| input_image=img_path, | |
| negative_prompt="", | |
| seed=42, | |
| upscale_factor=upscale_factor, | |
| controlnet_scale=controlnet_scale, | |
| controlnet_decay=controlnet_decay, | |
| condition_scale=condition_scale, | |
| tile_width=tile_width, | |
| tile_height=tile_height, | |
| denoise_strength=denoise_strength, | |
| num_inference_steps=num_inference_steps, | |
| solver=solver, | |
| model="finegrain/finegrain-image-enhancer", | |
| ) | |
| except Exception as e: | |
| raise gr.Error(f"Error in {e}") | |
| return result[0] | |
| async def gen( | |
| prompt:str, | |
| lora_add:str="", | |
| lora_word:str="", | |
| width:int=768, | |
| height:int=1024, | |
| scales:float=3.5, | |
| steps:int=24, | |
| seed:int=-1, | |
| upscale_factor:int=2, | |
| controlnet_scale:float=0.6, | |
| controlnet_decay:float=1, | |
| condition_scale:int=6, | |
| tile_width:int=112, | |
| tile_height:int=144, | |
| denoise_strength:float=0.35, | |
| num_inference_steps:int=18, | |
| solver:str="DDIM", | |
| progress=gr.Progress(track_tqdm=True) | |
| ): | |
| model = enable_lora(lora_add) | |
| print(model) | |
| image, seed = await generate_image(prompt, model, lora_word, width, height, scales, steps, seed) | |
| upscale_img = await upscale_image(prompt, image, upscale_factor, controlnet_scale, controlnet_decay, condition_scale, tile_width, tile_height, denoise_strength, num_inference_steps, solver) | |
| return image, upscale_img, seed | |
| with gr.Blocks(css=CSS, js=JS, theme="Nymbo/Nymbo_Theme") as demo: | |
| gr.HTML("<h1><center>Flux Lab Light</center></h1>") | |
| gr.HTML("<p><center>Powered By HF Inference API</center></p>") | |
| with gr.Row(): | |
| with gr.Column(scale=4): | |
| with gr.Row(): | |
| img = gr.Image(type="filepath", label='Flux Image', height=600) | |
| upscale_img = gr.Image(type="filepath", label='Upscale Image', height=600) | |
| with gr.Row(): | |
| prompt = gr.Textbox(label='Enter Your Prompt (Multi-Languages)', placeholder="Enter prompt...", scale=6) | |
| sendBtn = gr.Button(scale=1, variant='primary') | |
| with gr.Accordion("Advanced Options", open=True): | |
| with gr.Column(scale=1): | |
| width = gr.Slider( | |
| label="Width", | |
| minimum=512, | |
| maximum=1280, | |
| step=8, | |
| value=768, | |
| ) | |
| height = gr.Slider( | |
| label="Height", | |
| minimum=512, | |
| maximum=1280, | |
| step=8, | |
| value=1024, | |
| ) | |
| scales = gr.Slider( | |
| label="Guidance", | |
| minimum=3.5, | |
| maximum=7, | |
| step=0.1, | |
| value=3.5, | |
| ) | |
| steps = gr.Slider( | |
| label="Steps", | |
| minimum=1, | |
| maximum=100, | |
| step=1, | |
| value=24, | |
| ) | |
| seed = gr.Slider( | |
| label="Seeds", | |
| minimum=-1, | |
| maximum=MAX_SEED, | |
| step=1, | |
| value=-1, | |
| ) | |
| lora_add = gr.Textbox( | |
| label="Add Flux LoRA", | |
| info="Copy the HF LoRA model name here", | |
| lines=1, | |
| value="XLabs-AI/flux-RealismLora" | |
| ) | |
| lora_word = gr.Textbox( | |
| label="Add Flux LoRA Trigger Word", | |
| info="Add the Trigger Word", | |
| lines=1, | |
| value="", | |
| ) | |
| upscale_factor = gr.Radio( | |
| label="UpScale Factor", | |
| choices=[ | |
| 2, 3, 4 | |
| ], | |
| value=2, | |
| scale=2 | |
| ) | |
| controlnet_scale = gr.Slider( | |
| label="ControlNet Scale", | |
| minimum=0.1, | |
| maximum=1.0, | |
| step=0.1, | |
| value=0.6 | |
| ) | |
| controlnet_decay = gr.Slider( | |
| label="ControlNet Decay", | |
| minimum=0.1, | |
| maximum=1.0, | |
| step=0.1, | |
| value=1 | |
| ) | |
| condition_scale = gr.Slider( | |
| label="Condition Scale", | |
| minimum=1, | |
| maximum=10, | |
| step=1, | |
| value=6 | |
| ) | |
| tile_width = gr.Slider( | |
| label="Tile Width", | |
| minimum=64, | |
| maximum=256, | |
| step=16, | |
| value=112 | |
| ) | |
| tile_height = gr.Slider( | |
| label="Tile Height", | |
| minimum=64, | |
| maximum=256, | |
| step=16, | |
| value=144 | |
| ) | |
| denoise_strength = gr.Slider( | |
| label="Denoise Strength", | |
| minimum=0.1, | |
| maximum=1.0, | |
| step=0.1, | |
| value=0.35 | |
| ) | |
| num_inference_steps = gr.Slider( | |
| label="Num Inference Steps", | |
| minimum=1, | |
| maximum=50, | |
| step=1, | |
| value=18 | |
| ) | |
| solver = gr.Radio( | |
| label="Solver", | |
| choices=[ | |
| "DDIM", "DPM" | |
| ], | |
| value="DDIM", | |
| scale=2 | |
| ) | |
| gr.on( | |
| triggers=[ | |
| prompt.submit, | |
| sendBtn.click, | |
| ], | |
| fn=gen, | |
| inputs=[ | |
| prompt, | |
| lora_add, | |
| lora_word, | |
| width, | |
| height, | |
| scales, | |
| steps, | |
| seed, | |
| upscale_factor, | |
| controlnet_scale, | |
| controlnet_decay, | |
| condition_scale, | |
| tile_width, | |
| tile_height, | |
| denoise_strength, | |
| num_inference_steps, | |
| solver | |
| ], | |
| outputs=[img, upscale_img, seed] | |
| ) | |
| if __name__ == "__main__": | |
| demo.queue(api_open=False).launch(show_api=False, share=False) |