import gradio as gr import requests import io import random import os from PIL import Image from deep_translator import GoogleTranslator API_URL = "https://api-inference.huggingface.co/models/openskyml/dalle-3-xl" API_TOKEN = os.getenv("HF_READ_TOKEN") headers = {"Authorization": f"Bearer {API_TOKEN}"} timeout = 100 models_list = ["AbsoluteReality 1.8.1", "DALL-E 3 XL", "Playground 2", "Openjourney 4", "Lyriel 1.6", "Animagine XL 2.0", "Counterfeit 2.5", "Realistic Vision 5.1", "Incursios 1.6", "Anime Detailer XL", "Vector Art XL", "epiCRealism", "PixelArt XL", "NewReality XL", "Anything 5.0", "Disney", "CleanLinearMix", "OrangeMixs"] # PLEASE ❤ like ❤ this space. Please like me. I am 12 years old, one of my projects is: https://ai-hub.rf.gd . I live in Russia, I don't know English very well. Therefore, I apologize that there is only Russian here, but I think it will not be difficult to translate all this. (For example, using gpt) def query(prompt, model, is_negative=False, steps=30, cfg_scale=7, sampler="DPM++ 2M Karras", seed=-1): if prompt == "" or prompt == None: return None API_TOKEN = random.choice([os.getenv("HF_READ_TOKEN"), os.getenv("HF_READ_TOKEN_2"), os.getenv("HF_READ_TOKEN_3"), os.getenv("HF_READ_TOKEN_4"), os.getenv("HF_READ_TOKEN_5")]) # it is free headers = {"Authorization": f"Bearer {API_TOKEN}"} key = random.randint(0, 999) prompt = GoogleTranslator(source='ru', target='en').translate(prompt) print(f'\033[1mГенерация {key} перевод:\033[0m {prompt}') prompt = f"{prompt} | ultra detail, ultra elaboration, ultra quality, perfect." print(f'\033[1mГенерация {key}:\033[0m {prompt}') if model == 'DALL-E 3 XL': API_URL = "https://api-inference.huggingface.co/models/openskyml/dalle-3-xl" if model == 'Playground 2': API_URL = "https://api-inference.huggingface.co/models/playgroundai/playground-v2-1024px-aesthetic" if model == 'Openjourney 4': API_URL = "https://api-inference.huggingface.co/models/prompthero/openjourney-v4" if model == 'AbsoluteReality 1.8.1': API_URL = "https://api-inference.huggingface.co/models/digiplay/AbsoluteReality_v1.8.1" if model == 'Lyriel 1.6': API_URL = "https://api-inference.huggingface.co/models/stablediffusionapi/lyrielv16" if model == 'Animagine XL 2.0': API_URL = "https://api-inference.huggingface.co/models/Linaqruf/animagine-xl-2.0" prompt = f"Anime. {prompt}" if model == 'Counterfeit 2.5': API_URL = "https://api-inference.huggingface.co/models/gsdf/Counterfeit-V2.5" if model == 'Realistic Vision 5.1': API_URL = "https://api-inference.huggingface.co/models/stablediffusionapi/realistic-vision-v51" if model == 'Incursios 1.6': API_URL = "https://api-inference.huggingface.co/models/digiplay/incursiosMemeDiffusion_v1.6" if model == 'Anime Detailer XL': API_URL = "https://api-inference.huggingface.co/models/Linaqruf/anime-detailer-xl-lora" prompt = f"Anime. {prompt}" if model == 'epiCRealism': API_URL = "https://api-inference.huggingface.co/models/emilianJR/epiCRealism" if model == 'PixelArt XL': API_URL = "https://api-inference.huggingface.co/models/nerijs/pixel-art-xl" if model == 'NewReality XL': API_URL = "https://api-inference.huggingface.co/models/stablediffusionapi/newrealityxl-global-nsfw" if model == 'Anything 5.0': API_URL = "https://api-inference.huggingface.co/models/hogiahien/anything-v5-edited" if model == 'Vector Art XL': API_URL = "https://api-inference.huggingface.co/models/DoctorDiffusion/doctor-diffusion-s-controllable-vector-art-xl-lora" if model == 'Disney': API_URL = "https://api-inference.huggingface.co/models/goofyai/disney_style_xl" prompt = f"Disney style. {prompt}" if model == 'CleanLinearMix': API_URL = "https://api-inference.huggingface.co/models/digiplay/CleanLinearMix_nsfw" if model == 'OrangeMixs': API_URL = "https://api-inference.huggingface.co/models/WarriorMama777/OrangeMixs" payload = { "inputs": prompt, "is_negative": is_negative, "steps": steps, "cfg_scale": cfg_scale, "seed": seed if seed != -1 else random.randint(1, 1000000000) } response = requests.post(API_URL, headers=headers, json=payload, timeout=timeout) if response.status_code != 200: print(f"Ошибка: Не удалось получить изображение. Статус ответа: {response.status_code}") print(f"Содержимое ответа: {response.text}") return None try: image_bytes = response.content image = Image.open(io.BytesIO(image_bytes)) print(f'\033[1mГенерация {key} завершена!\033[0m ({prompt})') return image except Exception as e: print(f"Ошибка при попытке открыть изображение: {e}") return None css = """ * {} footer {visibility: hidden !important;} """ with gr.Blocks(css=css) as dalle: with gr.Tab("Базовые настройки"): with gr.Row(): with gr.Column(elem_id="prompt-container"): with gr.Row(): text_prompt = gr.Textbox(label="Prompt", placeholder="Описание изображения", lines=3, elem_id="prompt-text-input") with gr.Row(): model = gr.Radio(label="Модель", value="DALL-E 3 XL", choices=models_list) with gr.Tab("Расширенные настройки"): with gr.Row(): negative_prompt = gr.Textbox(label="Negative Prompt", placeholder="Чего не должно быть на изображении", value="[deformed | disfigured], poorly drawn, [bad : wrong] anatomy, [extra | missing | floating | disconnected] limb, (mutated hands and fingers), blurry, text, fuzziness", lines=3, elem_id="negative-prompt-text-input") with gr.Row(): steps = gr.Slider(label="Sampling steps", value=35, minimum=1, maximum=100, step=1) with gr.Row(): cfg = gr.Slider(label="CFG Scale", value=7, minimum=1, maximum=20, step=1) with gr.Row(): method = gr.Radio(label="Sampling method", value="DPM++ 2M Karras", choices=["DPM++ 2M Karras", "DPM++ SDE Karras", "Euler", "Euler a", "Heun", "DDIM"]) with gr.Row(): seed = gr.Slider(label="Seed", value=-1, minimum=-1, maximum=1000000000, step=1) with gr.Tab("Информация"): with gr.Row(): gr.Textbox(label="Шаблон prompt", value="{prompt} | ultra detail, ultra elaboration, ultra quality, perfect.") with gr.Row(): text_button = gr.Button("Генерация", variant='primary', elem_id="gen-button") with gr.Row(): image_output = gr.Image(type="pil", label="Изображение", elem_id="gallery") text_button.click(query, inputs=[text_prompt, model, negative_prompt, steps, cfg, method, seed], outputs=image_output) dalle.launch(show_api=False, share=False)