Update utils.py

app.py

@@ -6,15 +6,12 @@ from stablepy import (
     SCHEDULE_PREDICTION_TYPE_OPTIONS,
     check_scheduler_compatibility,
     TASK_AND_PREPROCESSORS,
-    FACE_RESTORATION_MODELS,
-    scheduler_names,
 )
 from constants import (
     DIRECTORY_MODELS,
     DIRECTORY_LORAS,
     DIRECTORY_VAES,
     DIRECTORY_EMBEDS,
-    DIRECTORY_UPSCALERS,
     DOWNLOAD_MODEL,
     DOWNLOAD_VAE,
     DOWNLOAD_LORA,
@@ -38,12 +35,15 @@ from constants import (
     EXAMPLES_GUI,
     RESOURCES,
     DIFFUSERS_CONTROLNET_MODEL,
-    IP_MODELS,
-    MODE_IP_OPTIONS,
 )
 from stablepy.diffusers_vanilla.style_prompt_config import STYLE_NAMES
 import torch
 import re
+from stablepy import (
+    scheduler_names,
+    IP_ADAPTERS_SD,
+    IP_ADAPTERS_SDXL,
+)
 import time
 from PIL import ImageFile
 from utils import (
@@ -70,15 +70,13 @@ import warnings
 from stablepy import logger
 from diffusers import FluxPipeline
 # import urllib.parse
-import subprocess
 
-subprocess.run("rm -rf /data-nvme/zerogpu-offload/*", env={}, shell=True)
 ImageFile.LOAD_TRUNCATED_IMAGES = True
 torch.backends.cuda.matmul.allow_tf32 = True
 # os.environ["PYTORCH_NO_CUDA_MEMORY_CACHING"] = "1"
 print(os.getenv("SPACES_ZERO_GPU"))
 
-directories = [DIRECTORY_MODELS, DIRECTORY_LORAS, DIRECTORY_VAES, DIRECTORY_EMBEDS, DIRECTORY_UPSCALERS]
+directories = [DIRECTORY_MODELS, DIRECTORY_LORAS, DIRECTORY_VAES, DIRECTORY_EMBEDS]
 for directory in directories:
     os.makedirs(directory, exist_ok=True)
 
@@ -121,8 +119,8 @@ flux_pipe = FluxPipeline.from_pretrained(
     torch_dtype=torch.bfloat16,
 ).to("cuda")
 components = flux_pipe.components
+components.pop("transformer", None)
 delete_model(flux_repo)
-# components = None
 
 #######################
 # GUI
@@ -175,7 +173,6 @@ class GuiSD:
         dtype_model = torch.bfloat16 if model_type == "FLUX" else torch.float16
 
         if not os.path.exists(model_name):
-            print("debug", model_name, vae_model, task, controlnet_model)
             _ = download_diffuser_repo(
                 repo_name=model_name,
                 model_type=model_type,
@@ -201,7 +198,10 @@ class GuiSD:
         yield f"Loading model: {model_name}"
 
         if vae_model == "BakedVAE":
-            vae_model = model_name
+            if not os.path.exists(model_name):
+                vae_model = model_name
+            else:
+                vae_model = None
         elif vae_model:
             vae_type = "SDXL" if "sdxl" in vae_model.lower() else "SD 1.5"
             if model_type != vae_type:
@@ -310,8 +310,8 @@ class GuiSD:
         syntax_weights,
         upscaler_model_path,
         upscaler_increases_size,
-        upscaler_tile_size,
-        upscaler_tile_overlap,
+        esrgan_tile,
+        esrgan_tile_overlap,
         hires_steps,
         hires_denoising_strength,
         hires_sampler,
@@ -375,9 +375,6 @@ class GuiSD:
         mode_ip2,
         scale_ip2,
         pag_scale,
-        face_restoration_model,
-        face_restoration_visibility,
-        face_restoration_weight,
     ):
         info_state = html_template_message("Navigating latent space...")
         yield info_state, gr.update(), gr.update()
@@ -416,20 +413,23 @@ class GuiSD:
         self.model.stream_config(concurrency=concurrency, latent_resize_by=1, vae_decoding=False)
 
         if task != "txt2img" and not image_control:
-            raise ValueError("Reference image is required. Please upload one in 'Image ControlNet/Inpaint/Img2img'.")
+            raise ValueError("No control image found: To use this function, you have to upload an image in 'Image ControlNet/Inpaint/Img2img'")
 
-        if task in ["inpaint", "repaint"] and not image_mask:
-            raise ValueError("Mask image not found. Upload one in 'Image Mask' to proceed.")
+        if task == "inpaint" and not image_mask:
+            raise ValueError("No mask image found: Specify one in 'Image Mask'")
 
-        if "https://" not in str(UPSCALER_DICT_GUI[upscaler_model_path]):
+        if upscaler_model_path in UPSCALER_KEYS[:9]:
             upscaler_model = upscaler_model_path
         else:
+            directory_upscalers = 'upscalers'
+            os.makedirs(directory_upscalers, exist_ok=True)
+
             url_upscaler = UPSCALER_DICT_GUI[upscaler_model_path]
 
-            if not os.path.exists(f"./{DIRECTORY_UPSCALERS}/{url_upscaler.split('/')[-1]}"):
-                download_things(DIRECTORY_UPSCALERS, url_upscaler, HF_TOKEN)
+            if not os.path.exists(f"./upscalers/{url_upscaler.split('/')[-1]}"):
+                download_things(directory_upscalers, url_upscaler, HF_TOKEN)
 
-            upscaler_model = f"./{DIRECTORY_UPSCALERS}/{url_upscaler.split('/')[-1]}"
+            upscaler_model = f"./upscalers/{url_upscaler.split('/')[-1]}"
 
         logging.getLogger("ultralytics").setLevel(logging.INFO if adetailer_verbose else logging.ERROR)
 
@@ -531,8 +531,8 @@ class GuiSD:
             "t2i_adapter_conditioning_factor": float(t2i_adapter_conditioning_factor),
             "upscaler_model_path": upscaler_model,
             "upscaler_increases_size": upscaler_increases_size,
-            "upscaler_tile_size": upscaler_tile_size,
-            "upscaler_tile_overlap": upscaler_tile_overlap,
+            "esrgan_tile": esrgan_tile,
+            "esrgan_tile_overlap": esrgan_tile_overlap,
             "hires_steps": hires_steps,
             "hires_denoising_strength": hires_denoising_strength,
             "hires_prompt": hires_prompt,
@@ -547,9 +547,6 @@ class GuiSD:
             "ip_adapter_model": params_ip_model,
             "ip_adapter_mode": params_ip_mode,
             "ip_adapter_scale": params_ip_scale,
-            "face_restoration_model": face_restoration_model,
-            "face_restoration_visibility": face_restoration_visibility,
-            "face_restoration_weight": face_restoration_weight,
         }
 
         # kwargs for diffusers pipeline
@@ -697,26 +694,22 @@ def sd_gen_generate_pipeline(*args):
 
 
 @spaces.GPU(duration=15)
-def process_upscale(image, upscaler_name, upscaler_size):
+def esrgan_upscale(image, upscaler_name, upscaler_size):
     if image is None: return None
 
     from stablepy.diffusers_vanilla.utils import save_pil_image_with_metadata
-    from stablepy import load_upscaler_model
+    from stablepy import UpscalerESRGAN
 
-    image = image.convert("RGB")
     exif_image = extract_exif_data(image)
 
-    name_upscaler = UPSCALER_DICT_GUI[upscaler_name]
-
-    if "https://" in str(name_upscaler):
-
-        if not os.path.exists(f"./{DIRECTORY_UPSCALERS}/{name_upscaler.split('/')[-1]}"):
-            download_things(DIRECTORY_UPSCALERS, name_upscaler, HF_TOKEN)
+    url_upscaler = UPSCALER_DICT_GUI[upscaler_name]
+    directory_upscalers = 'upscalers'
+    os.makedirs(directory_upscalers, exist_ok=True)
+    if not os.path.exists(f"./upscalers/{url_upscaler.split('/')[-1]}"):
+        download_things(directory_upscalers, url_upscaler, HF_TOKEN)
 
-        name_upscaler = f"./{DIRECTORY_UPSCALERS}/{name_upscaler.split('/')[-1]}"
-
-    scaler_beta = load_upscaler_model(model=name_upscaler, tile=0, tile_overlap=8, device="cuda", half=True)
-    image_up = scaler_beta.upscale(image, upscaler_size, True)
+    scaler_beta = UpscalerESRGAN(0, 0)
+    image_up = scaler_beta.upscale(image, upscaler_size, f"./upscalers/{url_upscaler.split('/')[-1]}")
 
     image_path = save_pil_image_with_metadata(image_up, f'{os.getcwd()}/up_images', exif_image)
 
@@ -907,8 +900,8 @@ with gr.Blocks(theme="NoCrypt/miku", css=CSS) as app:
 
                     upscaler_model_path_gui = gr.Dropdown(label="Upscaler", choices=UPSCALER_KEYS, value=UPSCALER_KEYS[0])
                     upscaler_increases_size_gui = gr.Slider(minimum=1.1, maximum=4., step=0.1, value=1.2, label="Upscale by")
-                    upscaler_tile_size_gui = gr.Slider(minimum=0, maximum=512, step=16, value=0, label="Upscaler Tile Size", info="0 = no tiling")
-                    upscaler_tile_overlap_gui = gr.Slider(minimum=0, maximum=48, step=1, value=8, label="Upscaler Tile Overlap")
+                    esrgan_tile_gui = gr.Slider(minimum=0, value=0, maximum=500, step=1, label="ESRGAN Tile")
+                    esrgan_tile_overlap_gui = gr.Slider(minimum=1, maximum=200, step=1, value=8, label="ESRGAN Tile Overlap")
                     hires_steps_gui = gr.Slider(minimum=0, value=30, maximum=100, step=1, label="Hires Steps")
                     hires_denoising_strength_gui = gr.Slider(minimum=0.1, maximum=1.0, step=0.01, value=0.55, label="Hires Denoising Strength")
                     hires_sampler_gui = gr.Dropdown(label="Hires Sampler", choices=POST_PROCESSING_SAMPLER, value=POST_PROCESSING_SAMPLER[0])
@@ -957,16 +950,11 @@ with gr.Blocks(theme="NoCrypt/miku", css=CSS) as app:
                             [lora1_gui, lora2_gui, lora3_gui, lora4_gui, lora5_gui, lora6_gui, lora7_gui, new_lora_status]
                         )
 
-                with gr.Accordion("Face restoration", open=False, visible=True):
-
-                    face_rest_options = [None] + FACE_RESTORATION_MODELS
-
-                    face_restoration_model_gui = gr.Dropdown(label="Face restoration model", choices=face_rest_options, value=face_rest_options[0])
-                    face_restoration_visibility_gui = gr.Slider(minimum=0., maximum=1., step=0.001, value=1., label="Visibility")
-                    face_restoration_weight_gui = gr.Slider(minimum=0., maximum=1., step=0.001, value=.5, label="Weight", info="(0 = maximum effect, 1 = minimum effect)")
-
                 with gr.Accordion("IP-Adapter", open=False, visible=True):
 
+                    IP_MODELS = sorted(list(set(IP_ADAPTERS_SD + IP_ADAPTERS_SDXL)))
+                    MODE_IP_OPTIONS = ["original", "style", "layout", "style+layout"]
+
                     with gr.Accordion("IP-Adapter 1", open=False, visible=True):
                         image_ip1 = gr.Image(label="IP Image", type="filepath")
                         mask_ip1 = gr.Image(label="IP Mask", type="filepath")
@@ -985,13 +973,13 @@ with gr.Blocks(theme="NoCrypt/miku", css=CSS) as app:
                     image_mask_gui = gr.Image(label="Image Mask", type="filepath")
                     strength_gui = gr.Slider(
                         minimum=0.01, maximum=1.0, step=0.01, value=0.55, label="Strength",
-                        info="This option adjusts the level of changes for img2img, repaint and inpaint."
+                        info="This option adjusts the level of changes for img2img and inpainting."
                     )
                     image_resolution_gui = gr.Slider(
                         minimum=64, maximum=2048, step=64, value=1024, label="Image Resolution",
                         info="The maximum proportional size of the generated image based on the uploaded image."
                     )
-                    controlnet_model_gui = gr.Dropdown(label="ControlNet model", choices=DIFFUSERS_CONTROLNET_MODEL, value=DIFFUSERS_CONTROLNET_MODEL[0], allow_custom_value=True)
+                    controlnet_model_gui = gr.Dropdown(label="ControlNet model", choices=DIFFUSERS_CONTROLNET_MODEL, value=DIFFUSERS_CONTROLNET_MODEL[0])
                     control_net_output_scaling_gui = gr.Slider(minimum=0, maximum=5.0, step=0.1, value=1, label="ControlNet Output Scaling in UNet")
                     control_net_start_threshold_gui = gr.Slider(minimum=0, maximum=1, step=0.01, value=0, label="ControlNet Start Threshold (%)")
                     control_net_stop_threshold_gui = gr.Slider(minimum=0, maximum=1, step=0.01, value=1, label="ControlNet Stop Threshold (%)")
@@ -1202,11 +1190,8 @@ with gr.Blocks(theme="NoCrypt/miku", css=CSS) as app:
 
         with gr.Row():
             with gr.Column():
-
-                USCALER_TAB_KEYS = [name for name in UPSCALER_KEYS[9:]]
-
                 image_up_tab = gr.Image(label="Image", type="pil", sources=["upload"])
-                upscaler_tab = gr.Dropdown(label="Upscaler", choices=USCALER_TAB_KEYS, value=USCALER_TAB_KEYS[5])
+                upscaler_tab = gr.Dropdown(label="Upscaler", choices=UPSCALER_KEYS[9:], value=UPSCALER_KEYS[11])
                 upscaler_size_tab = gr.Slider(minimum=1., maximum=4., step=0.1, value=1.1, label="Upscale by")
                 generate_button_up_tab = gr.Button(value="START UPSCALE", variant="primary")
 
@@ -1214,7 +1199,7 @@ with gr.Blocks(theme="NoCrypt/miku", css=CSS) as app:
                 result_up_tab = gr.Image(label="Result", type="pil", interactive=False, format="png")
 
                 generate_button_up_tab.click(
-                    fn=process_upscale,
+                    fn=esrgan_upscale,
                     inputs=[image_up_tab, upscaler_tab, upscaler_size_tab],
                     outputs=[result_up_tab],
                 )
@@ -1286,8 +1271,8 @@ with gr.Blocks(theme="NoCrypt/miku", css=CSS) as app:
             prompt_syntax_gui,
             upscaler_model_path_gui,
             upscaler_increases_size_gui,
-            upscaler_tile_size_gui,
-            upscaler_tile_overlap_gui,
+            esrgan_tile_gui,
+            esrgan_tile_overlap_gui,
             hires_steps_gui,
             hires_denoising_strength_gui,
             hires_sampler_gui,
@@ -1351,9 +1336,6 @@ with gr.Blocks(theme="NoCrypt/miku", css=CSS) as app:
             mode_ip2,
             scale_ip2,
             pag_scale_gui,
-            face_restoration_model_gui,
-            face_restoration_visibility_gui,
-            face_restoration_weight_gui,
             load_lora_cpu_gui,
             verbose_info_gui,
             gpu_duration_gui,

constants.py

@@ -4,9 +4,6 @@ from stablepy import (
     scheduler_names,
     SD15_TASKS,
     SDXL_TASKS,
-    ALL_BUILTIN_UPSCALERS,
-    IP_ADAPTERS_SD,
-    IP_ADAPTERS_SDXL,
 )
 
 # - **Download Models**
@@ -21,7 +18,6 @@ DOWNLOAD_LORA = "https://huggingface.co/Leopain/color/resolve/main/Coloring_book
 LOAD_DIFFUSERS_FORMAT_MODEL = [
     'stabilityai/stable-diffusion-xl-base-1.0',
     'Laxhar/noobai-XL-1.1',
-    'Laxhar/noobai-XL-Vpred-1.0',
     'black-forest-labs/FLUX.1-dev',
     'John6666/blue-pencil-flux1-v021-fp8-flux',
     'John6666/wai-ani-flux-v10forfp8-fp8-flux',
@@ -33,8 +29,7 @@ LOAD_DIFFUSERS_FORMAT_MODEL = [
     'shauray/FluxDev-HyperSD-merged',
     'mikeyandfriends/PixelWave_FLUX.1-dev_03',
     'terminusresearch/FluxBooru-v0.3',
-    'black-forest-labs/FLUX.1-schnell',
-    # 'ostris/OpenFLUX.1',
+    'ostris/OpenFLUX.1',
     'shuttleai/shuttle-3-diffusion',
     'Laxhar/noobai-XL-1.0',
     'John6666/noobai-xl-nai-xl-epsilonpred10version-sdxl',
@@ -48,22 +43,12 @@ LOAD_DIFFUSERS_FORMAT_MODEL = [
     'John6666/ntr-mix-illustrious-xl-noob-xl-ntrmix35-sdxl',
     'John6666/ntr-mix-illustrious-xl-noob-xl-v777-sdxl',
     'John6666/ntr-mix-illustrious-xl-noob-xl-v777forlora-sdxl',
-    'John6666/ntr-mix-illustrious-xl-noob-xl-xi-sdxl',
-    'John6666/ntr-mix-illustrious-xl-noob-xl-xii-sdxl',
-    'John6666/ntr-mix-illustrious-xl-noob-xl-xiii-sdxl',
-    'John6666/mistoon-anime-v10illustrious-sdxl',
-    'John6666/hassaku-xl-illustrious-v10-sdxl',
-    'John6666/hassaku-xl-illustrious-v10style-sdxl',
     'John6666/haruki-mix-illustrious-v10-sdxl',
     'John6666/noobreal-v10-sdxl',
     'John6666/complicated-noobai-merge-vprediction-sdxl',
-    'Laxhar/noobai-XL-Vpred-0.9r',
-    'Laxhar/noobai-XL-Vpred-0.75s',
-    'Laxhar/noobai-XL-Vpred-0.75',
     'Laxhar/noobai-XL-Vpred-0.65s',
     'Laxhar/noobai-XL-Vpred-0.65',
     'Laxhar/noobai-XL-Vpred-0.6',
-    'John6666/cat-tower-noobai-xl-checkpoint-v14vpred-sdxl',
     'John6666/noobai-xl-nai-xl-vpred05version-sdxl',
     'John6666/noobai-fusion2-vpred-itercomp-v1-sdxl',
     'John6666/noobai-xl-nai-xl-vpredtestversion-sdxl',
@@ -73,7 +58,6 @@ LOAD_DIFFUSERS_FORMAT_MODEL = [
     'John6666/illustrious-pencil-xl-v200-sdxl',
     'John6666/obsession-illustriousxl-v21-sdxl',
     'John6666/obsession-illustriousxl-v30-sdxl',
-    'John6666/obsession-illustriousxl-v31-sdxl',
     'John6666/wai-nsfw-illustrious-v70-sdxl',
     'John6666/illustrious-pony-mix-v3-sdxl',
     'John6666/nova-anime-xl-illustriousv10-sdxl',
@@ -84,7 +68,6 @@ LOAD_DIFFUSERS_FORMAT_MODEL = [
     'John6666/meinaxl-v2-sdxl',
     'Eugeoter/artiwaifu-diffusion-2.0',
     'comin/IterComp',
-    'John6666/epicrealism-xl-vxiabeast-sdxl',
     'John6666/epicrealism-xl-v10kiss2-sdxl',
     'John6666/epicrealism-xl-v8kiss-sdxl',
     'misri/zavychromaxl_v80',
@@ -98,7 +81,6 @@ LOAD_DIFFUSERS_FORMAT_MODEL = [
     'John6666/ras-real-anime-screencap-v1-sdxl',
     'John6666/duchaiten-pony-xl-no-score-v60-sdxl',
     'John6666/mistoon-anime-ponyalpha-sdxl',
-    'John6666/mistoon-xl-copper-v20fast-sdxl',
     'John6666/ebara-mfcg-pony-mix-v12-sdxl',
     'John6666/t-ponynai3-v51-sdxl',
     'John6666/t-ponynai3-v65-sdxl',
@@ -117,18 +99,13 @@ LOAD_DIFFUSERS_FORMAT_MODEL = [
     'John6666/cyberrealistic-pony-v63-sdxl',
     'John6666/cyberrealistic-pony-v64-sdxl',
     'John6666/cyberrealistic-pony-v65-sdxl',
-    'John6666/cyberrealistic-pony-v7-sdxl',
     'GraydientPlatformAPI/realcartoon-pony-diffusion',
     'John6666/nova-anime-xl-pony-v5-sdxl',
     'John6666/autismmix-sdxl-autismmix-pony-sdxl',
     'John6666/aimz-dream-real-pony-mix-v3-sdxl',
-    'John6666/prefectious-xl-nsfw-v10-sdxl',
-    'GraydientPlatformAPI/iniverseponyRealGuofeng49',
     'John6666/duchaiten-pony-real-v11fix-sdxl',
     'John6666/duchaiten-pony-real-v20-sdxl',
     'John6666/duchaiten-pony-xl-no-score-v70-sdxl',
-    'Spestly/OdysseyXL-3.0',
-    'Spestly/OdysseyXL-4.0',
     'KBlueLeaf/Kohaku-XL-Zeta',
     'cagliostrolab/animagine-xl-3.1',
     'yodayo-ai/kivotos-xl-2.0',
@@ -143,9 +120,7 @@ LOAD_DIFFUSERS_FORMAT_MODEL = [
     'digiplay/darkphoenix3D_v1.1',
     'digiplay/BeenYouLiteL11_diffusers',
     'GraydientPlatformAPI/rev-animated2',
-    'myxlmynx/cyberrealistic_classic40',
-    'GraydientPlatformAPI/cyberreal6',
-    'GraydientPlatformAPI/cyberreal5',
+    'youknownothing/cyberrealistic_v50',
     'youknownothing/deliberate-v6',
     'GraydientPlatformAPI/deliberate-cyber3',
     'GraydientPlatformAPI/picx-real',
@@ -181,7 +156,6 @@ DIRECTORY_MODELS = 'models'
 DIRECTORY_LORAS = 'loras'
 DIRECTORY_VAES = 'vaes'
 DIRECTORY_EMBEDS = 'embedings'
-DIRECTORY_UPSCALERS = 'upscalers'
 
 CACHE_HF = "/home/user/.cache/huggingface/hub/"
 STORAGE_ROOT = "/home/user/"
@@ -210,21 +184,27 @@ TASK_STABLEPY = {
     'optical pattern ControlNet': 'pattern',
     'recolor ControlNet': 'recolor',
     'tile ControlNet': 'tile',
-    'repaint ControlNet': 'repaint',
 }
 
 TASK_MODEL_LIST = list(TASK_STABLEPY.keys())
 
 UPSCALER_DICT_GUI = {
     None: None,
-    **{bu: bu for bu in ALL_BUILTIN_UPSCALERS if bu not in ["HAT x4", "DAT x4", "DAT x3", "DAT x2", "SwinIR 4x"]},
-    # "RealESRGAN_x4plus": "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.1.0/RealESRGAN_x4plus.pth",
+    "Lanczos": "Lanczos",
+    "Nearest": "Nearest",
+    'Latent': 'Latent',
+    'Latent (antialiased)': 'Latent (antialiased)',
+    'Latent (bicubic)': 'Latent (bicubic)',
+    'Latent (bicubic antialiased)': 'Latent (bicubic antialiased)',
+    'Latent (nearest)': 'Latent (nearest)',
+    'Latent (nearest-exact)': 'Latent (nearest-exact)',
+    "RealESRGAN_x4plus": "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.1.0/RealESRGAN_x4plus.pth",
     "RealESRNet_x4plus": "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.1.1/RealESRNet_x4plus.pth",
-    # "RealESRGAN_x4plus_anime_6B": "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.2.4/RealESRGAN_x4plus_anime_6B.pth",
-    # "RealESRGAN_x2plus": "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.1/RealESRGAN_x2plus.pth",
-    # "realesr-animevideov3": "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-animevideov3.pth",
-    # "realesr-general-x4v3": "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-general-x4v3.pth",
-    # "realesr-general-wdn-x4v3": "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-general-wdn-x4v3.pth",
+    "RealESRGAN_x4plus_anime_6B": "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.2.4/RealESRGAN_x4plus_anime_6B.pth",
+    "RealESRGAN_x2plus": "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.1/RealESRGAN_x2plus.pth",
+    "realesr-animevideov3": "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-animevideov3.pth",
+    "realesr-general-x4v3": "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-general-x4v3.pth",
+    "realesr-general-wdn-x4v3": "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-general-wdn-x4v3.pth",
     "4x-UltraSharp": "https://huggingface.co/Shandypur/ESRGAN-4x-UltraSharp/resolve/main/4x-UltraSharp.pth",
     "4x_foolhardy_Remacri": "https://huggingface.co/FacehugmanIII/4x_foolhardy_Remacri/resolve/main/4x_foolhardy_Remacri.pth",
     "Remacri4xExtraSmoother": "https://huggingface.co/hollowstrawberry/upscalers-backup/resolve/main/ESRGAN/Remacri%204x%20ExtraSmoother.pth",
@@ -239,7 +219,6 @@ UPSCALER_KEYS = list(UPSCALER_DICT_GUI.keys())
 DIFFUSERS_CONTROLNET_MODEL = [
     "Automatic",
 
-    "brad-twinkl/controlnet-union-sdxl-1.0-promax",
     "xinsir/controlnet-union-sdxl-1.0",
     "xinsir/anime-painter",
     "Eugeoter/noob-sdxl-controlnet-canny",
@@ -262,6 +241,7 @@ DIFFUSERS_CONTROLNET_MODEL = [
     "r3gm/controlnet-recolor-sdxl-fp16",
     "r3gm/controlnet-openpose-twins-sdxl-1.0-fp16",
     "r3gm/controlnet-qr-pattern-sdxl-fp16",
+    "brad-twinkl/controlnet-union-sdxl-1.0-promax",
     "Yakonrus/SDXL_Controlnet_Tile_Realistic_v2",
     "TheMistoAI/MistoLine",
     "briaai/BRIA-2.3-ControlNet-Recoloring",
@@ -340,20 +320,6 @@ POST_PROCESSING_SAMPLER = ["Use same sampler"] + [
     name_s for name_s in scheduler_names if "Auto-Loader" not in name_s
 ]
 
-IP_MODELS = []
-ALL_IPA = sorted(set(IP_ADAPTERS_SD + IP_ADAPTERS_SDXL))
-
-for origin_name in ALL_IPA:
-    suffixes = []
-    if origin_name in IP_ADAPTERS_SD:
-        suffixes.append("sd1.5")
-    if origin_name in IP_ADAPTERS_SDXL:
-        suffixes.append("sdxl")
-    ref_name = f"{origin_name} ({'/'.join(suffixes)})"
-    IP_MODELS.append((ref_name, origin_name))
-
-MODE_IP_OPTIONS = ["original", "style", "layout", "style+layout"]
-
 SUBTITLE_GUI = (
     "### This demo uses [diffusers](https://github.com/huggingface/diffusers)"
     " to perform different tasks in image generation."
@@ -374,9 +340,7 @@ EXAMPLES_GUI_HELP = (
     3. ControlNet Canny SDXL
     4. Optical pattern (Optical illusion) SDXL
     5. Convert an image to a coloring drawing
-    6. V prediction model inference
-    7. V prediction model sd_embed variant inference
-    8. ControlNet OpenPose SD 1.5 and Latent upscale
+    6. ControlNet OpenPose SD 1.5 and Latent upscale
 
     - Different tasks can be performed, such as img2img or using the IP adapter, to preserve a person's appearance or a specific style based on an image.
     """
@@ -503,54 +467,6 @@ EXAMPLES_GUI = [
         35,
         False,
     ],
-    [
-        "[mochizuki_shiina], [syuri22], newest, reimu, solo, outdoors, water, flower, lantern",
-        "worst quality, normal quality, old, sketch,",
-        28,
-        7.0,
-        -1,
-        "None",
-        0.33,
-        "DPM 3M Ef",
-        1600,
-        1024,
-        "Laxhar/noobai-XL-Vpred-1.0",
-        "txt2img",
-        "color_image.png",  # img conttol
-        1024,  # img resolution
-        0.35,  # strength
-        1.0,  # cn scale
-        0.0,  # cn start
-        1.0,  # cn end
-        "Classic",
-        None,
-        30,
-        False,
-    ],
-    [
-        "[mochizuki_shiina], [syuri22], newest, multiple girls, 2girls, earrings, jewelry, gloves, purple eyes, black hair, looking at viewer, nail polish, hat, smile, open mouth, fingerless gloves, sleeveless, :d, upper body, blue eyes, closed mouth, black gloves, hands up, long hair, shirt, bare shoulders, white headwear, blush, black headwear, blue nails, upper teeth only, short hair, white gloves, white shirt, teeth, rabbit hat, star earrings, purple nails, pink hair, detached sleeves, fingernails, fake animal ears, animal hat, sleeves past wrists, black shirt, medium hair, fur trim, sleeveless shirt, turtleneck, long sleeves, rabbit ears, star \\(symbol\\)",
-        "worst quality, normal quality, old, sketch,",
-        28,
-        7.0,
-        -1,
-        "None",
-        0.33,
-        "DPM 3M Ef",
-        1600,
-        1024,
-        "Laxhar/noobai-XL-Vpred-1.0",
-        "txt2img",
-        "color_image.png",  # img conttol
-        1024,  # img resolution
-        0.35,  # strength
-        1.0,  # cn scale
-        0.0,  # cn start
-        1.0,  # cn end
-        "Classic-sd_embed",
-        None,
-        30,
-        False,
-    ],
     [
         "1girl,face,curly hair,red hair,white background,",
         "(worst quality:2),(low quality:2),(normal quality:2),lowres,watermark,",

requirements.txt

@@ -1,7 +1,5 @@
-stablepy==0.6.2
+git+https://github.com/R3gm/stablepy.git@a9fe2dc  # -b refactor_sampler_fix
 torch==2.2.0
 gdown
 opencv-python
-unidecode
-pydantic==2.10.6
-huggingface_hub==0.29.3
+unidecode

utils.py

@@ -62,7 +62,7 @@ class ModelInformation:
         self.download_url = json_data.get("downloadUrl", "")
         self.model_url = f"https://civitai.com/models/{self.model_id}?modelVersionId={self.model_version_id}"
         self.filename_url = next(
-            (v.get("name", "") for v in json_data.get("files", []) if str(self.model_version_id) in v.get("downloadUrl", "") and v.get("type", "Model") == "Model"), ""
+            (v.get("name", "") for v in reversed(json_data.get("files", [])) if str(self.model_version_id) in v.get("downloadUrl", "")), ""
         )
         self.filename_url = self.filename_url if self.filename_url else ""
         self.description = json_data.get("description", "")
@@ -300,7 +300,7 @@ def get_model_type(repo_id: str):
     default = "SD 1.5"
     try:
         if os.path.exists(repo_id):
-            tag, _, _, _ = checkpoint_model_type(repo_id)
+            tag = checkpoint_model_type(repo_id)
             return DIFFUSECRAFT_CHECKPOINT_NAME[tag]
         else:
             model = api.model_info(repo_id=repo_id, timeout=5.0)