modify the pyproject.toml file

Merge pull request #182 from haohaocreates/publish
Add Github Action for Publishing to Comfy Registry
2024-05-23 13:33:03 +08:00 · 2024-05-23 13:30:16 +08:00 · 2024-05-23 13:30:00 +08:00 · 2024-05-23 12:33:45 +08:00 · 2024-05-22 18:08:19 -04:00 · 2024-05-22 18:08:16 -04:00
29 changed files with 1385 additions and 308 deletions
@@ -0,0 +1,21 @@
+name: Publish to Comfy registry
+on:
+  workflow_dispatch:
+  push:
+    branches:
+      - main
+    paths:
+      - "pyproject.toml"
+
+jobs:
+  publish-node:
+    name: Publish Custom Node to registry
+    runs-on: ubuntu-latest
+    steps:
+      - name: Check out code
+        uses: actions/checkout@v4
+      - name: Publish Custom Node
+        uses: Comfy-Org/publish-node-action@main
+        with:
+          ## Add your own personal access token to your Github Repository secrets and reference it here.
+          personal_access_token: ${{ secrets.REGISTRY_ACCESS_TOKEN }}
@@ -33,6 +33,16 @@
 - 
 ## Changelog

+**v1.1.7**
+
+- Added `easy prompt` - Subject and light presets, maybe adjusted later
+- Added `easy icLightApply` - Light and shadow migration, Code based on [ComfyUI-IC-Light](https://github.com/huchenlei/ComfyUI-IC-Light)
+- Added `easy imageSplitGrid`
+- `easy kSamplerInpainting` added options such as different diffusion and brushnet in **additional** widget
+- Support for brushnet model loading - [ComfyUI-BrushNet](https://github.com/nullquant/ComfyUI-BrushNet)
+- Added `easy applyFooocusInpaint` - Replace FooocusInpaintLoader
+- Removed `easy fooocusInpaintLoader`
+
 **v1.1.6**

 - Added **alignYourSteps** to **schedulder** widget in all `easy preSampling` and `easy fullkSampler`
@@ -12,7 +12,7 @@

 **ComfyUI-Easy-Use** 是一个化繁为简的节点整合包, 在 [tinyterraNodes](https://github.com/TinyTerra/ComfyUI_tinyterraNodes) 的基础上进行延展，并针对了诸多主流的节点包做了整合与优化，以达到更快更方便使用ComfyUI的目的，在保证自由度的同时还原了本属于Stable Diffusion的极致畅快出图体验。

-<img src="https://raw.githubusercontent.com/yolain/yolain-comfyui-workflow/main/Docs/workflow_node_compare.png">
+[![ComfyUI-Yolain-Workflows](https://github.com/yolain/ComfyUI-Easy-Use/assets/73304135/9a3f54bc-a677-4bf1-a196-8845dd57c942)](https://github.com/yolain/ComfyUI-Yolain-Workflows)

 ## 特色介绍

@@ -23,10 +23,10 @@
 - 可多选的风格化提示词选择器，默认是Fooocus的样式json，可自定义json放在styles底下，samples文件夹里可放预览图(名称和name一致,图片文件名如有空格需转为下划线'_')
 - 加载器可开启A1111提示词风格模式，可重现与webui生成近乎相同的图像，需先安装 [ComfyUI_smZNodes](https://github.com/shiimizu/ComfyUI_smZNodes)
 - 可使用`easy latentNoisy`或`easy preSamplingNoiseIn`节点实现对潜空间的噪声注入
- 简化 SD1.x、SD2.x、SDXL、SVD、Zero123等流程 [示例参考](https://github.com/yolain/ComfyUI-Easy-Use?tab=readme-ov-file#StableDiffusion)
- 简化 Stable Cascade [示例参考](https://github.com/yolain/ComfyUI-Easy-Use?tab=readme-ov-file#StableCascade)
- 简化 Layer Diffuse [示例参考](https://github.com/yolain/ComfyUI-Easy-Use?tab=readme-ov-file#LayerDiffusion)， 首次使用您可能需要运行 `pip install -r requirements.txt` 安装所需依赖
- 简化 InstantID [示例参考](https://github.com/yolain/ComfyUI-Easy-Use?tab=readme-ov-file#InstantID), 需先保证自定义节点包中安装了 [ComfyUI_InstantID](https://github.com/cubiq/ComfyUI_InstantID)
+- 简化 SD1.x、SD2.x、SDXL、SVD、Zero123等流程 
+- 简化 Stable Cascade [示例参考](https://github.com/yolain/ComfyUI-Yolain-Workflows?tab=readme-ov-file#1-13-stable-cascade)
+- 简化 Layer Diffuse [示例参考](https://github.com/yolain/ComfyUI-Yolain-Workflows?tab=readme-ov-file#2-3-layerdiffusion)
+- 简化 InstantID [示例参考](https://github.com/yolain/ComfyUI-Yolain-Workflows?tab=readme-ov-file#2-2-instantid), 需先保证自定义节点包中安装了 [ComfyUI_InstantID](https://github.com/cubiq/ComfyUI_InstantID)
 - 简化 IPAdapter, 需先保证自定义节点包中安装最新版v2的 [ComfyUI_IPAdapter_plus](https://github.com/cubiq/ComfyUI_IPAdapter_plus)
 - 扩展 XYplot 的可用性
 - 整合了Fooocus Inpaint功能
@@ -35,9 +35,22 @@
 - 支持BriaAI的RMBG-1.4模型的背景去除节点，[技术参考](https://huggingface.co/briaai/RMBG-1.4)
 - 支持 强制清理comfyUI模型显存占用
 - 支持Stable Diffusion 3 多账号API节点
+- 支持IC-Light的应用 [示例参考](https://github.com/yolain/ComfyUI-Yolain-Workflows?tab=readme-ov-file#2-5-ic-light) | [代码整合来源](https://github.com/huchenlei/ComfyUI-IC-Light) | [技术参考](https://github.com/lllyasviel/IC-Light)

 ## 更新日志

+**v1.1.7**
+
+- 修复 一些模型(如controlnet模型等)未成功写入缓存，导致修改前置节点束参数（如提示词）需要二次载入模型的问题
+- 增加 `easy prompt` - 主体和光影预置项，后期可能会调整
+- 增加 `easy icLightApply` - 重绘光影, 从[ComfyUI-IC-Light](https://github.com/huchenlei/ComfyUI-IC-Light)优化
+- 增加 `easy imageSplitGrid` - 图像网格拆分
+- `easy kSamplerInpainting` 的 **additional** 属性增加差异扩散和brushnet等相关选项 
+- 增加 brushnet模型加载的支持 - [ComfyUI-BrushNet](https://github.com/nullquant/ComfyUI-BrushNet)
+- 增加 `easy applyFooocusInpaint` - Fooocus内补节点 替代原有的 FooocusInpaintLoader
+- 移除 `easy fooocusInpaintLoader` - 容易bug，不再使用
+- 修改 easy kSampler等采样器中并联的model 不再替换输出中pipe里的model
+
 **v1.1.6**

 - 增加步调齐整适配 - 在所有的预采样和全采样器节点中的 调度器(schedulder) 增加了 **alignYourSteps** 选项
@@ -124,7 +137,7 @@

 - 修复未安装 ComfyUI-Impack-Pack 和 ComfyUI_InstantID 时报错
 - 修复 `easy pipeIn` - pipe设为可不必选
- 增加 `easy instantIDApply` - 需要先安装 [ComfyUI_InstantID](https://github.com/cubiq/ComfyUI_InstantID), 工作流参考[示例](https://github.com/yolain/ComfyUI-Easy-Use?tab=readme-ov-file#InstantID)
+- 增加 `easy instantIDApply` - 需要先安装 [ComfyUI_InstantID](https://github.com/cubiq/ComfyUI_InstantID), 工作流参考[示例](https://github.com/yolain/ComfyUI-Yolain-Workflows?tab=readme-ov-file#2-2-instantid)
 - 修复 `easy detailerFix` 未添加到保存图片格式化扩展名可用节点列表
 - 修复 `easy XYInputs: PromptSR` 在替换负面提示词时报错
 </details>
@@ -328,39 +341,8 @@
 | easy preSamplingLayerDiffusion | [ComfyUI-layerdiffusion](https://github.com/huchenlei/ComfyUI-layerdiffusion) | LayeredDiffusionApply等  | 
 | easy dynamiCrafterLoader       | [ComfyUI-layerdiffusion](https://github.com/ExponentialML/ComfyUI_Native_DynamiCrafter) | Apply Dynamicrafter     | 
 | easy imageChooser              | [cg-image-picker](https://github.com/chrisgoringe/cg-image-picker) | Preview Chooser         | 
-| easy styleAlignedBatchAlign              | [style_aligned_comfy](https://github.com/chrisgoringe/cg-image-picker) | styleAlignedBatchAlign       | 
-
-## 示例
-
-导入后请自行更换您目录里的大模型
-
-### StableDiffusion
-#### 文生图
-
-<img src="https://raw.githubusercontent.com/yolain/yolain-comfyui-workflow/main/Workflows/Simple/text_to_image.png">
-
-#### 图生图+controlnet
-
-<img src="https://raw.githubusercontent.com/yolain/yolain-comfyui-workflow/main/Workflows/Simple/image_to_image_controlnet.png">
-
-#### InstantID
-
-<img src="https://raw.githubusercontent.com/yolain/yolain-comfyui-workflow/main/Workflows/Simple/instantID.png">
-
-### LayerDiffusion
-#### SD15
-<img src="https://raw.githubusercontent.com/yolain/yolain-comfyui-workflow/main/Workflows/Simple/layer_diffusion_sd15.png">
-
-#### SDXL
-<img src="https://raw.githubusercontent.com/yolain/yolain-comfyui-workflow/main/Workflows/Simple/layer_diffusion_example.png">
-
-### StableCascade
-#### 文生图
-<img src="https://raw.githubusercontent.com/yolain/yolain-comfyui-workflow/main/Workflows/StableCascade/text_to_image.png">
-
-#### 图生图
-<img src="https://raw.githubusercontent.com/yolain/yolain-comfyui-workflow/main/Workflows/StableCascade/image_to_image.png">
-
+| easy styleAlignedBatchAlign    | [style_aligned_comfy](https://github.com/chrisgoringe/cg-image-picker) | styleAlignedBatchAlign  | 
+| easy icLightApply              | [ComfyUI-IC-Light](https://github.com/huchenlei/ComfyUI-IC-Light) | ICLightApply等           |

 ## Credits

@@ -387,3 +369,5 @@
 [ComfyUI-Custom-Scripts](https://github.com/pythongosssss/ComfyUI-Custom-Scripts) - pyssss 小蛇🐍脚本

 [cg-image-picker](https://github.com/chrisgoringe/cg-image-picker) - 图片选择器
+
+[ComfyUI-BrushNet](https://github.com/nullquant/ComfyUI-BrushNet) - BrushNet 内补节点
@@ -1,4 +1,4 @@
-__version__ = "1.1.6"
+__version__ = "1.1.7"

 import os
 import glob
@@ -0,0 +1,33 @@
+import folder_paths
+import os
+def add_folder_path_and_extensions(folder_name, full_folder_paths, extensions):
+    for full_folder_path in full_folder_paths:
+        folder_paths.add_model_folder_path(folder_name, full_folder_path)
+    if folder_name in folder_paths.folder_names_and_paths:
+        current_paths, current_extensions = folder_paths.folder_names_and_paths[folder_name]
+        updated_extensions = current_extensions | extensions
+        folder_paths.folder_names_and_paths[folder_name] = (current_paths, updated_extensions)
+    else:
+        folder_paths.folder_names_and_paths[folder_name] = (full_folder_paths, extensions)
+
+image_suffixs = set([".jpg", ".jpeg", ".png", ".gif", ".webp", ".bmp", ".tiff", ".svg", ".ico", ".apng", ".tif", ".hdr", ".exr"])
+
+model_path = folder_paths.models_dir
+add_folder_path_and_extensions("ultralytics_bbox", [os.path.join(model_path, "ultralytics", "bbox")], folder_paths.supported_pt_extensions)
+add_folder_path_and_extensions("ultralytics_segm", [os.path.join(model_path, "ultralytics", "segm")], folder_paths.supported_pt_extensions)
+add_folder_path_and_extensions("ultralytics", [os.path.join(model_path, "ultralytics")], folder_paths.supported_pt_extensions)
+add_folder_path_and_extensions("mmdets_bbox", [os.path.join(model_path, "mmdets", "bbox")], folder_paths.supported_pt_extensions)
+add_folder_path_and_extensions("mmdets_segm", [os.path.join(model_path, "mmdets", "segm")], folder_paths.supported_pt_extensions)
+add_folder_path_and_extensions("mmdets", [os.path.join(model_path, "mmdets")], folder_paths.supported_pt_extensions)
+add_folder_path_and_extensions("sams", [os.path.join(model_path, "sams")], folder_paths.supported_pt_extensions)
+add_folder_path_and_extensions("onnx", [os.path.join(model_path, "onnx")], {'.onnx'})
+add_folder_path_and_extensions("instantid", [os.path.join(model_path, "instantid")], folder_paths.supported_pt_extensions)
+add_folder_path_and_extensions("layer_model", [os.path.join(model_path, "layer_model")], folder_paths.supported_pt_extensions)
+add_folder_path_and_extensions("rembg", [os.path.join(model_path, "rembg")], folder_paths.supported_pt_extensions)
+add_folder_path_and_extensions("ipadapter", [os.path.join(model_path, "ipadapter")], folder_paths.supported_pt_extensions)
+add_folder_path_and_extensions("dynamicrafter_models", [os.path.join(model_path, "dynamicrafter_models")], folder_paths.supported_pt_extensions)
+add_folder_path_and_extensions("mediapipe", [os.path.join(model_path, "mediapipe")], set(['.tflite','.pth']))
+add_folder_path_and_extensions("inpaint", [os.path.join(model_path, "inpaint")], folder_paths.supported_pt_extensions)
+
+add_folder_path_and_extensions("checkpoints_thumb", [os.path.join(model_path, "checkpoints")], image_suffixs)
+add_folder_path_and_extensions("loras_thumb", [os.path.join(model_path, "loras")], image_suffixs)
@@ -237,17 +237,17 @@ def encode_token_weights_g(model, token_weight_pairs):


 def encode_token_weights_l(model, token_weight_pairs):
-    l_out, _ = model.clip_l.encode_token_weights(token_weight_pairs)
-    return l_out, None
+    l_out, pooled = model.clip_l.encode_token_weights(token_weight_pairs)
+    return l_out, pooled


 def encode_token_weights(model, token_weight_pairs, encode_func):
    if model.layer_idx is not None:
        # 2016 [c2cb8e88] 及以上版本去除了sdxl clip的clip_layer方法
-        if compare_revision(2016):
-            model.cond_stage_model.set_clip_options({'layer': model.layer_idx})
-        else:
-            model.cond_stage_model.clip_layer(model.layer_idx)
+        # if compare_revision(2016):
+        model.cond_stage_model.set_clip_options({'layer': model.layer_idx})
+        # else:
+        # model.cond_stage_model.clip_layer(model.layer_idx)

    model_management.load_model_gpu(model.patcher)
    return encode_func(model.cond_stage_model, token_weight_pairs)
@@ -316,8 +316,8 @@ def advanced_encode(clip, text, token_normalization, weight_interpretation, w_ma
            embeddings_final, pooled = advanced_encode_from_tokens(tokenized['l'],
                                                                   token_normalization,
                                                                   weight_interpretation,
-                                                                   lambda x: (clip.encode_from_tokens({'l': x}), None),
-                                                                   w_max=w_max)
+                                                                   lambda x: encode_token_weights(clip, x, encode_token_weights_l),
+                                                                   w_max=w_max,return_pooled=True,)
            cond = [[embeddings_final, {"pooled_output": pooled}]]

        if conditioning is not None:
@@ -38,7 +38,7 @@ MAX_SEED_NUM = 1125899906842624

 RESOURCES_DIR = os.path.join(Path(__file__).parent.parent, "resources")

-# fooocus
+# inpaint
 INPAINT_DIR = os.path.join(folder_paths.models_dir, "inpaint")
 FOOOCUS_STYLES_DIR = os.path.join(Path(__file__).parent.parent, "styles")
 FOOOCUS_STYLES_SAMPLES = 'https://raw.githubusercontent.com/lllyasviel/Fooocus/main/sdxl_styles/samples/'
@@ -58,6 +58,24 @@ FOOOCUS_INPAINT_PATCH = {
        "model_url": "https://huggingface.co/lllyasviel/fooocus_inpaint/resolve/main/inpaint.fooocus.patch"
    },
 }
+BRUSHNET_MODELS = {
+    "random_mask": {
+        "sd1": {
+            "model_url": "https://huggingface.co/Kijai/BrushNet-fp16/resolve/main/brushnet_random_mask_fp16.safetensors"
+        },
+        "sdxl": {
+            "model_url": "https://huggingface.co/yolain/brushnet/resolve/main/brushnet_random_mask_sdxl.safetensors"
+        }
+    },
+    "segmentation_mask": {
+        "sd1": {
+            "model_url": "https://huggingface.co/Kijai/BrushNet-fp16/resolve/main/brushnet_segmentation_mask_fp16.safetensors"
+        },
+        "sdxl": {
+            "model_url": "https://huggingface.co/yolain/brushnet/resolve/main/brushnet_segmentation_mask_sdxl.safetensors"
+        }
+    }
+}

 # layerDiffuse
 LAYER_DIFFUSION_DIR = os.path.join(folder_paths.models_dir, "layer_model")
@@ -94,7 +112,7 @@ LAYER_DIFFUSION = {
        }
    },
    "Everything": {
-        "sd15": {
+        "sd1": {
          "model_url": "https://huggingface.co/LayerDiffusion/layerdiffusion-v1/resolve/main/layer_sd15_joint.safetensors"
        },
        "sdxl": {
@@ -135,6 +153,27 @@ LAYER_DIFFUSION = {
    },
 }

+# IC Light
+IC_LIGHT_MODELS = {
+    "Foreground": {
+        "sd1": {
+            "model_url": "https://huggingface.co/huchenlei/IC-Light-ldm/resolve/main/iclight_sd15_fc_unet_ldm.safetensors"
+        },
+        "sdxl": {
+            "model_url": None
+        }
+    },
+    "Foreground&Background": {
+        "sd1": {
+            "model_url": "https://huggingface.co/huchenlei/IC-Light-ldm/resolve/main/iclight_sd15_fbc_unet_ldm.safetensors"
+        },
+        "sdxl": {
+            "model_url": None
+        }
+    }
+}
+
+
 # REMBG
 REMBG_DIR = os.path.join(folder_paths.models_dir, "rembg")
 REMBG_MODELS = {
@@ -1,6 +1,7 @@
-import sys, os, re, json, time, math, copy
+import sys, os, re, json, time
 import torch
 import folder_paths
+import numpy as np
 import comfy.utils, comfy.sample, comfy.samplers, comfy.controlnet, comfy.model_base, comfy.model_management
 try:
    import comfy.sampler_helpers
@@ -9,20 +10,21 @@ except:
 from comfy.sd import CLIP, VAE
 from comfy.model_patcher import ModelPatcher
 from comfy_extras.chainner_models import model_loading
-from comfy_extras.nodes_mask import LatentCompositeMasked
+from comfy_extras.nodes_mask import LatentCompositeMasked, GrowMask
+from comfy_extras.nodes_compositing import JoinImageWithAlpha
 from comfy.clip_vision import load as load_clip_vision
 from urllib.request import urlopen
 from PIL import Image

 from server import PromptServer
 from nodes import MAX_RESOLUTION, LatentFromBatch, RepeatLatentBatch, NODE_CLASS_MAPPINGS as ALL_NODE_CLASS_MAPPINGS, ConditioningSetMask, ConditioningConcat, CLIPTextEncode, VAEEncodeForInpaint, InpaintModelConditioning
-from .config import MAX_SEED_NUM, BASE_RESOLUTIONS, RESOURCES_DIR, INPAINT_DIR, FOOOCUS_STYLES_DIR, FOOOCUS_INPAINT_HEAD, FOOOCUS_INPAINT_PATCH, IPADAPTER_DIR, IPADAPTER_MODELS, DYNAMICRAFTER_DIR, DYNAMICRAFTER_MODELS
+from .config import MAX_SEED_NUM, BASE_RESOLUTIONS, RESOURCES_DIR, INPAINT_DIR, FOOOCUS_STYLES_DIR, FOOOCUS_INPAINT_HEAD, FOOOCUS_INPAINT_PATCH, BRUSHNET_MODELS, IPADAPTER_DIR, IPADAPTER_MODELS, DYNAMICRAFTER_DIR, DYNAMICRAFTER_MODELS, IC_LIGHT_MODELS
 from .log import log_node_info, log_node_error, log_node_warn
 from .wildcards import process_with_loras, get_wildcard_list, process
 from .adv_encode import advanced_encode
 from .layer_diffuse.func import LayerDiffuse, LayerMethod

-from .libs.utils import find_wildcards_seed, is_linked_styles_selector, easySave, get_local_filepath, add_folder_path_and_extensions, AlwaysEqualProxy, get_sd_version
+from .libs.utils import find_wildcards_seed, is_linked_styles_selector, easySave, get_local_filepath, AlwaysEqualProxy, get_sd_version
 from .libs.loader import easyLoader
 from .libs.sampler import easySampler, alignYourStepsScheduler
 from .libs.xyplot import easyXYPlot
@@ -33,29 +35,6 @@ from .libs.easing import EasingBase

 sampler = easySampler()
 easyCache = easyLoader()
-default_calculate_weight = copy.copy(ModelPatcher.calculate_weight)
-
-image_suffixs = set([".jpg", ".jpeg", ".png", ".gif", ".webp", ".bmp", ".tiff", ".svg", ".ico", ".apng", ".tif", ".hdr", ".exr"])
-
-model_path = folder_paths.models_dir
-add_folder_path_and_extensions("ultralytics_bbox", [os.path.join(model_path, "ultralytics", "bbox")], folder_paths.supported_pt_extensions)
-add_folder_path_and_extensions("ultralytics_segm", [os.path.join(model_path, "ultralytics", "segm")], folder_paths.supported_pt_extensions)
-add_folder_path_and_extensions("ultralytics", [os.path.join(model_path, "ultralytics")], folder_paths.supported_pt_extensions)
-add_folder_path_and_extensions("mmdets_bbox", [os.path.join(model_path, "mmdets", "bbox")], folder_paths.supported_pt_extensions)
-add_folder_path_and_extensions("mmdets_segm", [os.path.join(model_path, "mmdets", "segm")], folder_paths.supported_pt_extensions)
-add_folder_path_and_extensions("mmdets", [os.path.join(model_path, "mmdets")], folder_paths.supported_pt_extensions)
-add_folder_path_and_extensions("sams", [os.path.join(model_path, "sams")], folder_paths.supported_pt_extensions)
-add_folder_path_and_extensions("onnx", [os.path.join(model_path, "onnx")], {'.onnx'})
-add_folder_path_and_extensions("instantid", [os.path.join(model_path, "instantid")], folder_paths.supported_pt_extensions)
-add_folder_path_and_extensions("layer_model", [os.path.join(model_path, "layer_model")], folder_paths.supported_pt_extensions)
-add_folder_path_and_extensions("rembg", [os.path.join(model_path, "rembg")], folder_paths.supported_pt_extensions)
-add_folder_path_and_extensions("ipadapter", [os.path.join(model_path, "ipadapter")], folder_paths.supported_pt_extensions)
-add_folder_path_and_extensions("dynamicrafter_models", [os.path.join(model_path, "dynamicrafter_models")], folder_paths.supported_pt_extensions)
-add_folder_path_and_extensions("mediapipe", [os.path.join(model_path, "mediapipe")], set(['.tflite','.pth']))
-
-add_folder_path_and_extensions("checkpoints_thumb", [os.path.join(model_path, "checkpoints")], image_suffixs)
-add_folder_path_and_extensions("loras_thumb", [os.path.join(model_path, "loras")], image_suffixs)
-
 # ---------------------------------------------------------------提示词 开始----------------------------------------------------------------------#

 # 正面提示词
@@ -233,11 +212,64 @@ class stylesPromptSelector:
            positive_prompt = positive + ', '

        # 去重
-        positive_prompt = self.replace_repeat(positive_prompt) if positive_prompt else ''
-        negative_prompt = self.replace_repeat(negative_prompt) if negative_prompt else ''
+        # positive_prompt = self.replace_repeat(positive_prompt) if positive_prompt else ''
+        # negative_prompt = self.replace_repeat(negative_prompt) if negative_prompt else ''

        return (positive_prompt, negative_prompt)

+#prompt
+class prompt:
+    @classmethod
+    def INPUT_TYPES(s):
+        return {"required": {
+            "prompt": ("STRING", {"default": "", "multiline": True, "placeholder": "Prompt"}),
+            "main": ([
+                'none',
+                'beautiful woman, detailed face',
+                'handsome man, detailed face',
+                'pretty girl',
+                'handsome boy',
+                'dog',
+                'cat',
+                'Buddha',
+                'toy'
+            ], {"default": "none"}),
+            "lighting": ([
+                'none',
+                'sunshine from window',
+                'neon light, city',
+                'sunset over sea',
+                'golden time',
+                'sci-fi RGB glowing, cyberpunk',
+                'natural lighting',
+                'warm atmosphere, at home, bedroom',
+                'magic lit',
+                'evil, gothic, Yharnam',
+                'light and shadow',
+                'shadow from window',
+                'soft studio lighting',
+                'home atmosphere, cozy bedroom illumination',
+                'neon, Wong Kar-wai, warm',
+                'cinemative lighting',
+                'neo punk lighting, cyberpunk',
+            ],{"default":'none'})
+        }}
+
+    RETURN_TYPES = ("STRING",)
+    RETURN_NAMES = ("prompt",)
+    FUNCTION = "doit"
+
+    CATEGORY = "EasyUse/Prompt"
+
+    def doit(self, prompt, main, lighting):
+        if lighting != 'none' and main != 'none':
+            prompt = main + ',' + lighting + ',' + prompt
+        elif lighting != 'none' and main == 'none':
+            prompt = prompt + ',' + lighting
+        elif main != 'none':
+            prompt = main + ',' + prompt
+
+        return prompt,
 #promptList
 class promptList:
    @classmethod
@@ -1910,7 +1942,7 @@ class loraStackLoader:
                loras.append((lora_name, model_strength, clip_strength))
        return (loras,)

-class controlnetNameStack:
+class controlnetStack:

    def get_file_list(filenames):
        return [file for file in filenames if file != "put_models_here.txt" and "lllite" not in file]
@@ -1927,7 +1959,7 @@ class controlnetNameStack:
                "start_percent_1": ("FLOAT", {"default": 0.0, "min": 0.0, "max": 1.0, "step": 0.001}),
                "end_percent_1": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0, "step": 0.001}),
                "switch_2": (["Off", "On"],),
-                "controlnet_2": (s.controlnets,),
+                "`controlnet`_2": (s.controlnets,),
                "controlnet_strength_2": ("FLOAT", {"default": 1.0, "min": -10.0, "max": 10.0, "step": 0.01}),
                "start_percent_2": ("FLOAT", {"default": 0.0, "min": 0.0, "max": 1.0, "step": 0.001}),
                "end_percent_2": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0, "step": 0.001}),
@@ -1971,7 +2003,7 @@ class controlnetSimple:

    def controlnetApply(self, pipe, image, control_net_name, control_net=None, strength=1, scale_soft_weights=1):

-        positive, negative = easyControlnet().apply(control_net_name, image, pipe["positive"], pipe["negative"], strength, 0, 1, control_net, scale_soft_weights)
+        positive, negative = easyControlnet().apply(control_net_name, image, pipe["positive"], pipe["negative"], strength, 0, 1, control_net, scale_soft_weights, None, easyCache)

        new_pipe = {
            "model": pipe['model'],
@@ -2023,7 +2055,7 @@ class controlnetAdvanced:

    def controlnetApply(self, pipe, image, control_net_name, control_net=None, strength=1, start_percent=0, end_percent=1, scale_soft_weights=1):
        positive, negative = easyControlnet().apply(control_net_name, image, pipe["positive"], pipe["negative"],
-                                                    strength, start_percent, end_percent, control_net, scale_soft_weights)
+                                                    strength, start_percent, end_percent, control_net, scale_soft_weights, None, easyCache)

        new_pipe = {
            "model": pipe['model'],
@@ -2087,22 +2119,26 @@ class LLLiteLoader:
 # FooocusInpaint
 from .fooocus import InpaintHead, InpaintWorker
 inpaint_head_model = None
-class fooocusInpaintLoader:
+
+class applyFooocusInpaint:
    @classmethod
    def INPUT_TYPES(s):
        return {
            "required": {
+                "model": ("MODEL",),
+                "latent": ("LATENT",),
                "head": (list(FOOOCUS_INPAINT_HEAD.keys()),),
                "patch": (list(FOOOCUS_INPAINT_PATCH.keys()),),
-            }
+            },
        }

-    RETURN_TYPES = ("INPAINT_PATCH",)
-    RETURN_NAMES = ("patch",)
+    RETURN_TYPES = ("MODEL",)
+    RETURN_NAMES = ("model",)
    CATEGORY = "EasyUse/Inpaint"
    FUNCTION = "apply"

-    def apply(self, head, patch):
+    def apply(self, model, latent, head, patch):
+
        global inpaint_head_model

        head_file = get_local_filepath(FOOOCUS_INPAINT_HEAD[head]["model_url"], INPAINT_DIR)
@@ -2114,9 +2150,17 @@ class fooocusInpaintLoader:
        patch_file = get_local_filepath(FOOOCUS_INPAINT_PATCH[patch]["model_url"], INPAINT_DIR)
        inpaint_lora = comfy.utils.load_torch_file(patch_file, safe_load=True)

-        return ((inpaint_head_model, inpaint_lora),)
+        patch = (inpaint_head_model, inpaint_lora)
+        worker = InpaintWorker(node_name="easy kSamplerInpainting")
+        cloned = model.clone()
+
+        m, = worker.patch(cloned, latent, patch)
+        return (m,)
+

 #---------------------------------------------------------------适配器 开始----------------------------------------------------------------------#
+
+# 风格对齐
 from .libs.styleAlign import styleAlignBatch, SHARE_NORM_OPTIONS, SHARE_ATTN_OPTIONS
 class styleAlignedBatchAlign:

@@ -2138,6 +2182,92 @@ class styleAlignedBatchAlign:
    def align(self, model, share_norm, share_attn, scale):
        return (styleAlignBatch(model, share_norm, share_attn, scale),)

+# 光照对齐
+from .ic_light.func import ICLight, VAEEncodeArgMax
+class icLightApply:
+
+    @classmethod
+    def INPUT_TYPES(cls):
+        return {
+            "required": {
+                "mode": (list(IC_LIGHT_MODELS.keys()),),
+                "model": ("MODEL",),
+                "image": ("IMAGE",),
+                "vae": ("VAE",),
+                "lighting": (['None', 'Left Light', 'Right Light', 'Top Light', 'Bottom Light', 'Circle Light'],{"default": "None"}),
+                "source": (['Use Background Image', 'Use Flipped Background Image', 'Left Light', 'Right Light', 'Top Light', 'Bottom Light', 'Ambient'],{"default": "Use Background Image"}),
+                "remove_bg": ("BOOLEAN", {"default": True}),
+            },
+        }
+
+    RETURN_TYPES = ("MODEL", "IMAGE")
+    RETURN_NAMES = ("model", "lighting_image")
+    FUNCTION = "apply"
+    OUTPUT_NODE = True
+    CATEGORY = "EasyUse/Adapter"
+
+    def batch(self, image1, image2):
+        if image1.shape[1:] != image2.shape[1:]:
+            image2 = comfy.utils.common_upscale(image2.movedim(-1, 1), image1.shape[2], image1.shape[1], "bilinear",
+                                                "center").movedim(1, -1)
+        s = torch.cat((image1, image2), dim=0)
+        return s
+
+    def removebg(self, image):
+        if "easy imageRemBg" not in ALL_NODE_CLASS_MAPPINGS:
+            raise Exception("Please re-install ComfyUI-Easy-Use")
+        cls = ALL_NODE_CLASS_MAPPINGS['easy imageRemBg']
+        results = cls().remove('RMBG-1.4', image, 'Hide', 'ComfyUI')
+        if "result" in results:
+            image, _ = results['result']
+            return image
+
+    def apply(self, mode, model, image, vae, lighting, source, remove_bg):
+        model_type = get_sd_version(model)
+        if model_type == 'sdxl':
+            raise Exception("IC Light model is not supported for SDXL now")
+
+        batch_size, height, width, channel = image.shape
+        if channel == 3:
+            # remove bg
+            if mode == 'Foreground' or batch_size == 1:
+                if remove_bg:
+                    image = self.removebg(image)
+                else:
+                    mask = torch.full((1, height, width), 1.0, dtype=torch.float32, device="cpu")
+                    image, = JoinImageWithAlpha().join_image_with_alpha(image, mask)
+
+        iclight = ICLight()
+        if mode == 'Foreground':
+          lighting_image = iclight.generate_lighting_image(image, lighting)
+        else:
+          lighting_image = iclight.generate_source_image(image, source)
+          if source != 'Use Background Image':
+              _, height, width, _ = lighting_image.shape
+              mask = torch.full((1, height, width), 1.0, dtype=torch.float32, device="cpu")
+              lighting_image, = JoinImageWithAlpha().join_image_with_alpha(lighting_image, mask)
+              if batch_size < 2:
+                image = self.batch(image, lighting_image)
+              else:
+                original_image = [img.unsqueeze(0) for img in image]
+                original_image = self.removebg(original_image[0])
+                image = self.batch(original_image, lighting_image)
+
+        latent, = VAEEncodeArgMax().encode(vae, image)
+        key = 'iclight_' + mode + '_' + model_type
+        model_path = get_local_filepath(IC_LIGHT_MODELS[mode]['sd1']["model_url"],
+                                        os.path.join(folder_paths.models_dir, "unet"))
+        ic_model = None
+        if key in backend_cache.cache:
+            log_node_info("easy icLightApply", f"Using icLightModel {mode+'_'+model_type} Cached")
+            _, ic_model = backend_cache.cache[key][1]
+            m, _ = iclight.apply(model_path, model, latent, ic_model)
+        else:
+            m, ic_model = iclight.apply(model_path, model, latent, ic_model)
+            backend_cache.update_cache(key, 'iclight', (False, ic_model))
+        return (m, lighting_image)
+
+
 def insightface_loader(provider):
    try:
        from insightface.app import FaceAnalysis
@@ -2396,7 +2526,7 @@ class ipadapterApply(ipadapter):
                "weight_faceidv2": ("FLOAT", { "default": 1.0, "min": -1, "max": 5.0, "step": 0.05 }),
                "start_at": ("FLOAT", {"default": 0.0, "min": 0.0, "max": 1.0, "step": 0.001}),
                "end_at": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0, "step": 0.001}),
-                "cache_mode": (["insightface only", "clip_vision only", "ipadapter only", "all", "none"], {"default": "insightface only"},),
+                "cache_mode": (["insightface only", "clip_vision only", "ipadapter only", "all", "none"], {"default": "all"},),
                "use_tiled": ("BOOLEAN", {"default": False},),
            },

@@ -2458,7 +2588,7 @@ class ipadapterApplyAdvanced(ipadapter):
                "start_at": ("FLOAT", {"default": 0.0, "min": 0.0, "max": 1.0, "step": 0.001}),
                "end_at": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0, "step": 0.001}),
                "embeds_scaling": (['V only', 'K+V', 'K+V w/ C penalty', 'K+mean(V) w/ C penalty'],),
-                "cache_mode": (["insightface only", "clip_vision only","ipadapter only", "all", "none"], {"default": "insightface only"},),
+                "cache_mode": (["insightface only", "clip_vision only","ipadapter only", "all", "none"], {"default": "all"},),
                "use_tiled": ("BOOLEAN", {"default": False},),
                "use_batch": ("BOOLEAN", {"default": False},),
                "sharpening": ("FLOAT", {"default": 0.0, "min": 0.0, "max": 1.0, "step": 0.05}),
@@ -2527,7 +2657,7 @@ class ipadapterStyleComposition(ipadapter):
                "end_at": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0, "step": 0.001}),
                "embeds_scaling": (['V only', 'K+V', 'K+V w/ C penalty', 'K+mean(V) w/ C penalty'],),
                "cache_mode": (["insightface only", "clip_vision only", "ipadapter only", "all", "none"],
-                               {"default": "insightface only"},),
+                               {"default": "all"},),
            },
            "optional": {
                "image_composition": ("IMAGE",),
@@ -2849,7 +2979,8 @@ class instantID:
        # Apply InstantID
        if "ApplyInstantID" in ALL_NODE_CLASS_MAPPINGS:
            instantid_apply = ALL_NODE_CLASS_MAPPINGS['ApplyInstantID']
-            control_net = easyControlnet().load_controlnet(control_net_name, control_net, cn_soft_weights)
+            if control_net is None:
+                control_net = easyCache.load_controlnet(control_net_name, cn_soft_weights)
            model, positive, negative = instantid_apply().apply_instantid(instantid_model, insightface_model, control_net, image, model, positive, negative, start_at, end_at, weight=weight, ip_weight=None, cn_strength=cn_strength, noise=noise, image_kps=image_kps, mask=mask)
        else:
            self.error()
@@ -3005,9 +3136,14 @@ class samplerSettings:
        vae = pipe["vae"]
        batch_size = pipe["loader_settings"]["batch_size"] if "batch_size" in pipe["loader_settings"] else 1
        if image_to_latent is not None:
-            samples = {"samples": vae.encode(image_to_latent[:, :, :, :3])}
-            samples = RepeatLatentBatch().repeat(samples, batch_size)[0]
-            images = image_to_latent
+            _, height, width, _ = image_to_latent.shape
+            if height == 1 and width == 1:
+                samples = pipe["samples"]
+                images = pipe["images"]
+            else:
+                samples = {"samples": vae.encode(image_to_latent[:, :, :, :3])}
+                samples = RepeatLatentBatch().repeat(samples, batch_size)[0]
+                images = image_to_latent
        elif latent is not None:
            samples = latent
            images = pipe["images"]
@@ -3081,9 +3217,14 @@ class samplerSettingsAdvanced:
        vae = pipe["vae"]
        batch_size = pipe["loader_settings"]["batch_size"] if "batch_size" in pipe["loader_settings"] else 1
        if image_to_latent is not None:
-            samples = {"samples": vae.encode(image_to_latent[:, :, :, :3])}
-            samples = RepeatLatentBatch().repeat(samples, batch_size)[0]
-            images = image_to_latent
+            _, height, width, _ = image_to_latent.shape
+            if height == 1 and width == 1:
+                samples = pipe["samples"]
+                images = pipe["images"]
+            else:
+                samples = {"samples": vae.encode(image_to_latent[:, :, :, :3])}
+                samples = RepeatLatentBatch().repeat(samples, batch_size)[0]
+                images = image_to_latent
        elif latent is not None:
            samples = latent
            images = pipe["images"]
@@ -3186,7 +3327,6 @@ class samplerSettingsNoiseIn:

    def expand_mask(self, mask, expand, tapered_corners):
        try:
-            import numpy as np
            import scipy

            c = 0 if tapered_corners else 1
@@ -3398,13 +3538,18 @@ class samplerCustomSettings:
        batch_size = pipe["loader_settings"]["batch_size"] if "batch_size" in pipe["loader_settings"] else 1
        _guider, sigmas = None, None
        if image_to_latent is not None:
-            if guider == "IP2P+DualCFG":
-                positive, negative, latent = self.ip2p(pipe['positive'], pipe['negative'], vae, image_to_latent)
-                samples = latent
+            _, height, width, _ = image_to_latent.shape
+            if height == 1 and width == 1:
+                samples = pipe["samples"]
+                images = pipe["images"]
            else:
-                samples = {"samples": vae.encode(image_to_latent[:, :, :, :3])}
-                samples = RepeatLatentBatch().repeat(samples, batch_size)[0]
-            images = image_to_latent
+                if guider == "IP2P+DualCFG":
+                    positive, negative, latent = self.ip2p(pipe['positive'], pipe['negative'], vae, image_to_latent)
+                    samples = latent
+                else:
+                    samples = {"samples": vae.encode(image_to_latent[:, :, :, :3])}
+                    samples = RepeatLatentBatch().repeat(samples, batch_size)[0]
+                images = image_to_latent
        elif latent is not None:
            if guider == "IP2P+DualCFG":
                positive, negative, latent = self.ip2p(pipe['positive'], pipe['negative'], latent=latent)
@@ -4052,12 +4197,9 @@ class samplerFull(LayerDiffuse):
    FUNCTION = "run"
    CATEGORY = "EasyUse/Sampler"

-    def run(self, pipe, steps, cfg, sampler_name, scheduler, denoise, image_output, link_id, save_prefix, seed=None, model=None, positive=None, negative=None, latent=None, vae=None, clip=None, xyPlot=None, tile_size=None, prompt=None, extra_pnginfo=None, my_unique_id=None, force_full_denoise=False, disable_noise=False, downscale_options=None):
+    def run(self, pipe, steps, cfg, sampler_name, scheduler, denoise, image_output, link_id, save_prefix, seed=None, model=None, positive=None, negative=None, latent=None, vae=None, clip=None, xyPlot=None, tile_size=None, prompt=None, extra_pnginfo=None, my_unique_id=None, force_full_denoise=False, disable_noise=False, downscale_options=None, image=None):

-        # Clean loaded_objects
-        easyCache.update_loaded_objects(prompt)
-
-        samp_model = model if model is not None else pipe["model"]
+        samp_model = model.clone() if model is not None else pipe["model"].clone()
        samp_positive = positive if positive is not None else pipe["positive"]
        samp_negative = negative if negative is not None else pipe["negative"]
        samp_samples = latent if latent is not None else pipe["samples"]
@@ -4078,6 +4220,9 @@ class samplerFull(LayerDiffuse):
        add_noise = pipe['loader_settings']['add_noise'] if 'add_noise' in pipe['loader_settings'] else 'enabled'
        force_full_denoise = pipe['loader_settings']['force_full_denoise'] if 'force_full_denoise' in pipe['loader_settings'] else True

+        if image is not None and latent is None:
+            samp_samples = {"samples": samp_vae.encode(image[:, :, :, :3])}
+
        disable_noise = False
        if add_noise == "disable":
            disable_noise = True
@@ -4177,13 +4322,9 @@ class samplerFull(LayerDiffuse):
            spent_time = 'Diffusion:' + str((end_time-start_time)/1000)+'″, VAEDecode:' + str((end_decode_time-end_time)/1000)+'″ '

            results = easySave(new_images, save_prefix, image_output, prompt, extra_pnginfo)
-            sampler.update_value_by_id("results", my_unique_id, results)
-
-            # Clean loaded_objects
-            easyCache.update_loaded_objects(prompt)

            new_pipe = {
-                "model": samp_model,
+                **pipe,
                "positive": samp_positive,
                "negative": samp_negative,
                "vae": samp_vae,
@@ -4202,8 +4343,6 @@ class samplerFull(LayerDiffuse):
                }
            }

-            sampler.update_value_by_id("pipe_line", my_unique_id, new_pipe)
-
            del pipe

            if image_output == 'Preview&Choose':
@@ -4241,7 +4380,9 @@ class samplerFull(LayerDiffuse):
            if image_output in ("Sender", "Sender&Save"):
                PromptServer.instance.send_sync("img-send", {"link_id": link_id, "images": results})

-            ModelPatcher.calculate_weight = default_calculate_weight
+            if hasattr(ModelPatcher, "original_calculate_weight"):
+                ModelPatcher.calculate_weight = ModelPatcher.original_calculate_weight
+
            return {"ui": {"images": results},
                    "result": sampler.get_output(new_pipe,)}

@@ -4329,13 +4470,9 @@ class samplerFull(LayerDiffuse):
                                                                         output_images, samp_model)

            results = easySave(images, save_prefix, image_output, prompt, extra_pnginfo)
-            sampler.update_value_by_id("results", my_unique_id, results)
-
-            # Clean loaded_objects
-            easyCache.update_loaded_objects(prompt)

            new_pipe = {
-                "model": samp_model,
+                **pipe,
                "positive": samp_positive,
                "negative": samp_negative,
                "vae": samp_vae,
@@ -4351,14 +4488,14 @@ class samplerFull(LayerDiffuse):
                "loader_settings": pipe["loader_settings"],
            }

-            sampler.update_value_by_id("pipe_line", my_unique_id, new_pipe)
-
            del pipe

+            if hasattr(ModelPatcher, "original_calculate_weight"):
+                ModelPatcher.calculate_weight = ModelPatcher.original_calculate_weight
+
            if image_output in ("Hide", "Hide&Save"):
                return sampler.get_output(new_pipe)

-            ModelPatcher.calculate_weight = default_calculate_weight
            return {"ui": {"images": results}, "result": (sampler.get_output(new_pipe))}

        preview_latent = True
@@ -4442,6 +4579,7 @@ class samplerSimpleTiled:
    CATEGORY = "EasyUse/Sampler"

    def run(self, pipe, tile_size=512, image_output='preview', link_id=0, save_prefix='ComfyUI', model=None, prompt=None, extra_pnginfo=None, my_unique_id=None, force_full_denoise=False, disable_noise=False):
+
        return samplerFull().run(pipe, None, None,None,None,None, image_output, link_id, save_prefix,
                               None, model, None, None, None, None, None, None,
                               tile_size, prompt, extra_pnginfo, my_unique_id, force_full_denoise, disable_noise)
@@ -4477,6 +4615,7 @@ class samplerSimpleLayerDiffusion:
    CATEGORY = "EasyUse/Sampler"

    def run(self, pipe, image_output='preview', link_id=0, save_prefix='ComfyUI', model=None, prompt=None, extra_pnginfo=None, my_unique_id=None, force_full_denoise=False, disable_noise=False):
+
        result = samplerFull().run(pipe, None, None,None,None,None, image_output, link_id, save_prefix,
                               None, model, None, None, None, None, None, None,
                               None, prompt, extra_pnginfo, my_unique_id, force_full_denoise, disable_noise)
@@ -4563,12 +4702,11 @@ class samplerSimpleInpainting:
                 "image_output": (["Hide", "Preview", "Save", "Hide&Save", "Sender", "Sender&Save"],{"default": "Preview"}),
                 "link_id": ("INT", {"default": 0, "min": 0, "max": sys.maxsize, "step": 1}),
                 "save_prefix": ("STRING", {"default": "ComfyUI"}),
-                 "additional": (["None", "Differential Diffusion", "Only InpaintModelConditioning"],{"default": "None"})
+                 "additional": (["None", "InpaintModelCond", "Differential Diffusion", "Fooocus Inpaint", "Fooocus Inpaint + DD", "Brushnet Random", "Brushnet Random + DD", "Brushnet Segmentation", "Brushnet Segmentation + DD"],{"default": "None"})
                 },
                "optional": {
                    "model": ("MODEL",),
                    "mask": ("MASK",),
-                    "patch": ("INPAINT_PATCH",),
                },
                "hidden":
                  {"tile_size": "INT", "prompt": "PROMPT", "extra_pnginfo": "EXTRA_PNGINFO", "my_unique_id": "UNIQUE_ID",
@@ -4582,57 +4720,107 @@ class samplerSimpleInpainting:
    FUNCTION = "run"
    CATEGORY = "EasyUse/Sampler"

-    def run(self, pipe, grow_mask_by, image_output, link_id, save_prefix, additional, model=None, mask=None, patch=None, tile_size=None, prompt=None, extra_pnginfo=None, my_unique_id=None, force_full_denoise=False, disable_noise=False):
-        fooocus_model = None
-        model = model if model is not None else pipe['model']
+    def dd(self, model, positive, negative, pixels, vae, mask):
+        positive, negative, latent = InpaintModelConditioning().encode(positive, negative, pixels, vae, mask)
+        cls = ALL_NODE_CLASS_MAPPINGS['DifferentialDiffusion']
+        if cls is not None:
+            model, = cls().apply(model)
+        else:
+            raise Exception("Differential Diffusion not found,please update comfyui")
+        return positive, negative, latent, model
+
+    def get_brushnet_model(self, type, model):
+        model_type = 'sdxl' if isinstance(model.model.model_config, comfy.supported_models.SDXL) else 'sd1'
+        if type == 'random':
+            brush_model = BRUSHNET_MODELS['random_mask'][model_type]['model_url']
+            if model_type == 'sdxl':
+                pattern = 'brushnet.random.mask.sdxl.*\.(safetensors|bin)$'
+            else:
+                pattern = 'brushnet.random.mask.*\.(safetensors|bin)$'
+        elif type == 'segmentation':
+            brush_model = BRUSHNET_MODELS['segmentation_mask'][model_type]['model_url']
+            if model_type == 'sdxl':
+                pattern = 'brushnet.segmentation.mask.sdxl.*\.(safetensors|bin)$'
+            else:
+                pattern = 'brushnet.segmentation.mask.*\.(safetensors|bin)$'
+
+
+        brushfile = [e for e in folder_paths.get_filename_list('inpaint') if re.search(pattern, e, re.IGNORECASE)]
+        brushname = brushfile[0] if brushfile else None
+        if not brushname:
+            from urllib.parse import urlparse
+            get_local_filepath(brush_model, INPAINT_DIR)
+            parsed_url = urlparse(brush_model)
+            brushname = os.path.basename(parsed_url.path)
+        return brushname
+
+    def apply_brushnet(self, brushname, model, vae, image, mask, positive, negative, scale=1.0, start_at=0, end_at=10000):
+        if "BrushNetLoader" not in ALL_NODE_CLASS_MAPPINGS:
+            raise Exception("BrushNetLoader not found,please install ComfyUI-BrushNet")
+        cls = ALL_NODE_CLASS_MAPPINGS['BrushNetLoader']
+        brushnet, = cls().brushnet_loading(brushname, 'float16')
+        cls = ALL_NODE_CLASS_MAPPINGS['BrushNet']
+        m, positive, negative, latent = cls().model_update(model=model, vae=vae, image=image, mask=mask, brushnet=brushnet, positive=positive, negative=negative, scale=scale, start_at=start_at, end_at=end_at)
+        return m, positive, negative, latent
+
+    def run(self, pipe, grow_mask_by, image_output, link_id, save_prefix, additional, model=None, mask=None, tile_size=None, prompt=None, extra_pnginfo=None, my_unique_id=None, force_full_denoise=False, disable_noise=False):
+        _model = model if model is not None else pipe['model']
        latent = pipe['samples'] if 'samples' in pipe else None
        positive = pipe['positive']
        negative = pipe['negative']
-        pixels = pipe["images"] if pipe and "images" in pipe else None
+        images = pipe["images"] if pipe and "images" in pipe else None
        vae = pipe["vae"] if pipe and "vae" in pipe else None
        if 'noise_mask' in latent and mask is None:
            mask = latent['noise_mask']
-        else:
-            if pixels is None:
+        elif mask is not None:
+            if images is None:
                raise Exception("No Images found")
            if vae is None:
                raise Exception("No VAE found")
-            latent, = VAEEncodeForInpaint().encode(vae, pixels, mask, grow_mask_by)
-            mask = latent['noise_mask']

-        if mask is not None:
-            if additional != "None":
+        match additional:
+            case 'Differential Diffusion':
+                positive, negative, latent, _model = self.dd(_model, positive, negative, images, vae, mask)
+            case 'InpaintModelCond':
+                if mask is not None:
+                    mask, = GrowMask().expand_mask(mask, grow_mask_by, False)
+                positive, negative, latent = InpaintModelConditioning().encode(positive, negative, images, vae, mask)
+            case 'Fooocus Inpaint':
+                head = list(FOOOCUS_INPAINT_HEAD.keys())[0]
+                patch = list(FOOOCUS_INPAINT_PATCH.keys())[0]
+                if mask is not None:
+                    latent, = VAEEncodeForInpaint().encode(vae, images, mask, grow_mask_by)
+                _model, = applyFooocusInpaint().apply(_model, latent, head, patch)
+            case 'Fooocus Inpaint + DD':
+                head = list(FOOOCUS_INPAINT_HEAD.keys())[0]
+                patch = list(FOOOCUS_INPAINT_PATCH.keys())[0]
+                if mask is not None:
+                    latent, = VAEEncodeForInpaint().encode(vae, images, mask, grow_mask_by)
+                _model, = applyFooocusInpaint().apply(_model, latent, head, patch)
+                positive, negative, latent, _model = self.dd(_model, positive, negative, images, vae, mask)
+            case 'Brushnet Random':
+                mask, = GrowMask().expand_mask(mask, grow_mask_by, False)
+                brush_name = self.get_brushnet_model('random', _model)
+                _model, positive, negative, latent = self.apply_brushnet(brush_name, _model, vae, images, mask, positive, negative)
+            case 'Brushnet Random + DD':
+                mask, = GrowMask().expand_mask(mask, grow_mask_by, False)
+                brush_name = self.get_brushnet_model('random', _model)
+                _model, positive, negative, latent = self.apply_brushnet(brush_name, _model, vae, images, mask, positive, negative)
+                positive, negative, latent, _model = self.dd(_model, positive, negative, images, vae, mask)
+            case 'Brushnet Segmentation':
+                mask, = GrowMask().expand_mask(mask, grow_mask_by, False)
+                brush_name = self.get_brushnet_model('segmentation', _model)
+                _model, positive, negative, latent = self.apply_brushnet(brush_name, _model, vae, images, mask, positive, negative)
+            case 'Brushnet Segmentation + DD':
+                mask, = GrowMask().expand_mask(mask, grow_mask_by, False)
+                brush_name = self.get_brushnet_model('segmentation', _model)
+                _model, positive, negative, latent = self.apply_brushnet(brush_name, _model, vae, images, mask, positive, negative)
+                positive, negative, latent, _model = self.dd(_model, positive, negative, images, vae, mask)
+            case _:
+                latent, = VAEEncodeForInpaint().encode(vae, images, mask, grow_mask_by)

-                positive, negative, latent = InpaintModelConditioning().encode(positive, negative, pixels, vae, mask)
-                if additional == "Differential Diffusion":
-                    cls = ALL_NODE_CLASS_MAPPINGS['DifferentialDiffusion']
-                    if cls is not None:
-                        model, = cls().apply(model)
-                    else:
-                        raise Exception("Differential Diffusion not found,please update comfyui")
-
-            # when patch was linked
-            fooocus_model = None
-            if patch is not None:
-                worker = InpaintWorker(node_name="easy kSamplerInpainting")
-                fooocus_model, = worker.patch(model, latent, patch)
-
-            new_pipe = {
-                **pipe,
-                "model": fooocus_model if fooocus_model else model,
-                "positive": positive,
-                "negative": negative,
-                "vae": vae,
-                "samples": latent,
-                "loader_settings": pipe["loader_settings"],
-            }
-
-        else:
-            new_pipe = pipe
-        del pipe
-
-        results = samplerFull().run(new_pipe, None, None,None,None,None, image_output, link_id, save_prefix,
-                               None, None, None, None, None, None, None, None,
+        results = samplerFull().run(pipe, None, None,None,None,None, image_output, link_id, save_prefix,
+                               None, _model, positive, negative, latent, vae, None, None,
                               tile_size, prompt, extra_pnginfo, my_unique_id, force_full_denoise, disable_noise)

        result = results['result']
@@ -5596,8 +5784,7 @@ class detailerFix:
        del pipe

        if image_output in ("Hide", "Hide&Save"):
-            return {"ui": {},
-                    "result": (new_pipe, result_img, result_cropped_enhanced, result_cropped_enhanced_alpha, result_mask, result_cnet_images )}
+            return (new_pipe, result_img, result_cropped_enhanced, result_cropped_enhanced_alpha, result_mask, result_cnet_images)

        if image_output in ("Sender", "Sender&Save"):
            PromptServer.instance.send_sync("img-send", {"link_id": link_id, "images": results})
@@ -7033,6 +7220,7 @@ NODE_CLASS_MAPPINGS = {
    "easy positive": positivePrompt,
    "easy negative": negativePrompt,
    "easy wildcards": wildcardsPrompt,
+    "easy prompt": prompt,
    "easy promptList": promptList,
    "easy promptLine": promptLine,
    "easy promptConcat": promptConcat,
@@ -7063,8 +7251,10 @@ NODE_CLASS_MAPPINGS = {
    "easy instantIDApply": instantIDApply,
    "easy instantIDApplyADV": instantIDApplyAdvanced,
    "easy styleAlignedBatchAlign": styleAlignedBatchAlign,
+    "easy icLightApply": icLightApply,
    # Inpaint 内补
-    "easy fooocusInpaintLoader": fooocusInpaintLoader,
+    # "easy fooocusInpaintLoader": fooocusInpaintLoader,
+    "easy applyFooocusInpaint": applyFooocusInpaint,
    # latent 潜空间
    "easy latentNoisy": latentNoisy,
    "easy latentCompositeMaskedWithCond": latentCompositeMaskedWithCond,
@@ -7139,6 +7329,7 @@ NODE_DISPLAY_NAME_MAPPINGS = {
    "easy positive": "Positive",
    "easy negative": "Negative",
    "easy wildcards": "Wildcards",
+    "easy prompt": "Prompt",
    "easy promptList": "PromptList",
    "easy promptLine": "PromptLine",
    "easy promptConcat": "PromptConcat",
@@ -7169,8 +7360,10 @@ NODE_DISPLAY_NAME_MAPPINGS = {
    "easy instantIDApply": "Easy Apply InstantID",
    "easy instantIDApplyADV": "Easy Apply InstantID (Advanced)",
    "easy styleAlignedBatchAlign": "Easy Apply StyleAlign",
+    "easy icLightApply": "Easy Apply ICLight",
    # Inpaint 内补
-    "easy fooocusInpaintLoader": "Load Fooocus Inpaint",
+    # "easy fooocusInpaintLoader": "Load Fooocus Inpaint(Removed)",
+    "easy applyFooocusInpaint": "Apply Fooocus Inpaint",
    # latent 潜空间
    "easy latentNoisy": "LatentNoisy",
    "easy latentCompositeMaskedWithCond": "LatentCompositeMaskedWithCond",
@@ -19,6 +19,8 @@ class InpaintWorker:
    def __init__(self, node_name):
        self.node_name = node_name if node_name is not None else ""
        self.original_calculate_weight = ModelPatcher.calculate_weight
+        if not hasattr(ModelPatcher, "original_calculate_weight"):
+            ModelPatcher.original_calculate_weight = self.original_calculate_weight
        self.injected_model_patcher_calculate_weight = False

    def load_fooocus_patch(self, lora: dict, to_load: dict):
@@ -0,0 +1,185 @@
+import torch
+import numpy as np
+from typing import Tuple, TypedDict, Callable
+
+import comfy.model_management
+from comfy.sd import load_unet
+from comfy.ldm.models.autoencoder import AutoencoderKL
+from comfy.model_base import BaseModel
+from PIL import Image
+from nodes import VAEEncode
+
+from ..layer_diffuse.model import ModelPatcher, calculate_weight_adjust_channel
+from ..libs.image import np2tensor, pil2tensor
+
+class UnetParams(TypedDict):
+    input: torch.Tensor
+    timestep: torch.Tensor
+    c: dict
+    cond_or_uncond: torch.Tensor
+
+
+class VAEEncodeArgMax(VAEEncode):
+    def encode(self, vae, pixels):
+        assert isinstance(
+            vae.first_stage_model, AutoencoderKL
+        ), "ArgMax only supported for AutoencoderKL"
+        original_sample_mode = vae.first_stage_model.regularization.sample
+        vae.first_stage_model.regularization.sample = False
+        ret = super().encode(vae, pixels)
+        vae.first_stage_model.regularization.sample = original_sample_mode
+        return ret
+
+class ICLight:
+
+    @staticmethod
+    def apply_c_concat(params: UnetParams, concat_conds) -> UnetParams:
+        """Apply c_concat on unet call."""
+        sample = params["input"]
+        params["c"]["c_concat"] = torch.cat(
+            (
+                    [concat_conds.to(sample.device)]
+                    * (sample.shape[0] // concat_conds.shape[0])
+            ),
+            dim=0,
+        )
+        return params
+
+    @staticmethod
+    def create_custom_conv(
+        original_conv: torch.nn.Module,
+        dtype: torch.dtype,
+        device=torch.device,
+    ) -> torch.nn.Module:
+        with torch.no_grad():
+            new_conv_in = torch.nn.Conv2d(
+                8,
+                original_conv.out_channels,
+                original_conv.kernel_size,
+                original_conv.stride,
+                original_conv.padding,
+            )
+            new_conv_in.weight.zero_()
+            new_conv_in.weight[:, :4, :, :].copy_(original_conv.weight)
+            new_conv_in.bias = original_conv.bias
+            return new_conv_in.to(dtype=dtype, device=device)
+
+    def generate_lighting_image(self, original_image, direction):
+        _, image_height, image_width, _ = original_image.shape
+        match direction:
+            case 'Left Light':
+                gradient = np.linspace(255, 0, image_width)
+                image = np.tile(gradient, (image_height, 1))
+                input_bg = np.stack((image,) * 3, axis=-1).astype(np.uint8)
+                return np2tensor(input_bg)
+            case 'Right Light':
+                gradient = np.linspace(0, 255, image_width)
+                image = np.tile(gradient, (image_height, 1))
+                input_bg = np.stack((image,) * 3, axis=-1).astype(np.uint8)
+                return np2tensor(input_bg)
+            case 'Top Light':
+                gradient = np.linspace(255, 0, image_height)[:, None]
+                image = np.tile(gradient, (1, image_width))
+                input_bg = np.stack((image,) * 3, axis=-1).astype(np.uint8)
+                return np2tensor(input_bg)
+            case 'Bottom Light':
+                gradient = np.linspace(0, 255, image_height)[:, None]
+                image = np.tile(gradient, (1, image_width))
+                input_bg = np.stack((image,) * 3, axis=-1).astype(np.uint8)
+                return np2tensor(input_bg)
+            case 'Circle Light':
+                x = np.linspace(-1, 1, image_width)
+                y = np.linspace(-1, 1, image_height)
+                x, y = np.meshgrid(x, y)
+                r = np.sqrt(x ** 2 + y ** 2)
+                r = r / r.max()
+                color1 = np.array([0, 0, 0])[np.newaxis, np.newaxis, :]
+                color2 = np.array([255, 255, 255])[np.newaxis, np.newaxis, :]
+                gradient = (color1 * r[..., np.newaxis] + color2 * (1 - r)[..., np.newaxis]).astype(np.uint8)
+                image = pil2tensor(Image.fromarray(gradient))
+                return image
+            case _:
+                image = pil2tensor(Image.new('RGB', (1, 1), (0, 0, 0)))
+                return image
+
+    def generate_source_image(self, original_image, source):
+        batch_size, image_height, image_width, _ = original_image.shape
+        match source:
+            case 'Use Flipped Background Image':
+                if batch_size < 2:
+                    raise ValueError('Must be at least 2 image to use flipped background image.')
+                original_image = [img.unsqueeze(0) for img in original_image]
+                image = torch.flip(original_image[1], [2])
+                return image
+            case 'Ambient':
+                input_bg = np.zeros(shape=(image_height, image_width, 3), dtype=np.uint8) + 64
+                return np2tensor(input_bg)
+            case 'Left Light':
+                gradient = np.linspace(224, 32, image_width)
+                image = np.tile(gradient, (image_height, 1))
+                input_bg = np.stack((image,) * 3, axis=-1).astype(np.uint8)
+                return np2tensor(input_bg)
+            case 'Right Light':
+                gradient = np.linspace(32, 224, image_width)
+                image = np.tile(gradient, (image_height, 1))
+                input_bg = np.stack((image,) * 3, axis=-1).astype(np.uint8)
+                return np2tensor(input_bg)
+            case 'Top Light':
+                gradient = np.linspace(224, 32, image_height)[:, None]
+                image = np.tile(gradient, (1, image_width))
+                input_bg = np.stack((image,) * 3, axis=-1).astype(np.uint8)
+                return np2tensor(input_bg)
+            case 'Bottom Light':
+                gradient = np.linspace(32, 224, image_height)[:, None]
+                image = np.tile(gradient, (1, image_width))
+                input_bg = np.stack((image,) * 3, axis=-1).astype(np.uint8)
+                return np2tensor(input_bg)
+            case _:
+                image = pil2tensor(Image.new('RGB', (1, 1), (0, 0, 0)))
+                return image
+
+
+    def apply(self, ic_model_path, model: ModelPatcher, c_concat: dict, ic_model=None) -> Tuple[ModelPatcher]:
+        try:
+            ModelPatcher.calculate_weight = calculate_weight_adjust_channel(ModelPatcher.calculate_weight)
+        except:
+            pass
+
+        device = comfy.model_management.get_torch_device()
+        dtype = comfy.model_management.unet_dtype()
+        work_model = model.clone()
+
+        # Apply scale factor.
+        base_model: BaseModel = work_model.model
+        scale_factor = base_model.model_config.latent_format.scale_factor
+
+        # [B, 4, H, W]
+        concat_conds: torch.Tensor = c_concat["samples"] * scale_factor
+        # [1, 4 * B, H, W]
+        concat_conds = torch.cat([c[None, ...] for c in concat_conds], dim=1)
+
+        def unet_dummy_apply(unet_apply: Callable, params: UnetParams):
+            """A dummy unet apply wrapper serving as the endpoint of wrapper
+            chain."""
+            return unet_apply(x=params["input"], t=params["timestep"], **params["c"])
+
+        existing_wrapper = work_model.model_options.get(
+            "model_function_wrapper", unet_dummy_apply
+        )
+
+        def wrapper_func(unet_apply: Callable, params: UnetParams):
+            return existing_wrapper(unet_apply, params=self.apply_c_concat(params, concat_conds))
+
+        work_model.set_model_unet_function_wrapper(wrapper_func)
+        if not ic_model:
+            ic_model = load_unet(ic_model_path)
+        ic_model_state_dict = ic_model.model.diffusion_model.state_dict()
+
+        work_model.add_patches(
+            patches={
+                ("diffusion_model." + key): (value.to(dtype=dtype, device=device),)
+                for key, value in ic_model_state_dict.items()
+            }
+        )
+
+        return (work_model, ic_model)
@@ -1,4 +1,4 @@
-from PIL import Image
+from PIL import Image, ImageDraw, ImageFilter
 import os
 import hashlib
 import folder_paths
@@ -8,9 +8,10 @@ import comfy.model_management
 from comfy_extras.nodes_compositing import JoinImageWithAlpha
 from server import PromptServer
 from nodes import MAX_RESOLUTION
+from torchvision.transforms import Resize, CenterCrop, InterpolationMode
 from torchvision.transforms.functional import to_pil_image
 from .log import log_node_info
-from .libs.image import pil2tensor, tensor2pil, ResizeMode, get_new_bounds, RGB2RGBA, image2mask, blendImage
+from .libs.image import pil2tensor, tensor2pil, ResizeMode, get_new_bounds, RGB2RGBA, image2mask, mask2image, blendImage
 from .libs.colorfix import adain_color_fix, wavelet_color_fix
 from .libs.chooser import ChooserMessage, ChooserCancelled
 from .config import REMBG_DIR, REMBG_MODELS, HUMANPARSING_MODELS, MEDIAPIPE_MODELS, MEDIAPIPE_DIR
@@ -539,6 +540,41 @@ class imageSplitList:
          new_images[1].append(img)
    return new_images

+class imageSplitGrid:
+  @classmethod
+  def INPUT_TYPES(s):
+    return {
+      "required": {
+        "images": ("IMAGE",),
+        "row": ("INT", {"default": 1,"min": 1,"max": 10,"step": 1,}),
+        "column": ("INT", {"default": 1,"min": 1,"max": 10,"step": 1,}),
+      }
+    }
+
+  RETURN_TYPES = ("IMAGE",)
+  RETURN_NAMES = ("images",)
+  FUNCTION = "doit"
+  CATEGORY = "EasyUse/Image"
+
+  def crop(self, image, width, height, x, y):
+    x = min(x, image.shape[2] - 1)
+    y = min(y, image.shape[1] - 1)
+    to_x = width + x
+    to_y = height + y
+    img = image[:, y:to_y, x:to_x, :]
+    return img
+
+  def doit(self, images, row, column):
+    _, height, width, _ = images.shape
+    sub_width = width // column
+    sub_height = height // row
+    new_images = []
+    for i in range(row):
+        for j in range(column):
+          new_images.append(self.crop(images, sub_width, sub_height, j * sub_width, i * sub_height))
+
+    return (torch.cat(new_images, dim=0),)
+
 class imagesSplitImage:
    @classmethod
    def INPUT_TYPES(s):
@@ -557,6 +593,34 @@ class imagesSplitImage:
      new_images = torch.chunk(images, len(images), dim=0)
      return new_images

+
+class imageConcat:
+  @classmethod
+  def INPUT_TYPES(s):
+    return {"required": {
+      "image1": ("IMAGE",),
+      "image2": ("IMAGE",),
+      "direction": (['right','down','left','up',],{"default": 'right'}),
+      "match_image_size": ("BOOLEAN", {"default": False}),
+    }}
+
+  RETURN_TYPES = ("IMAGE",)
+  FUNCTION = "concat"
+  CATEGORY = "EasyUse/Image"
+
+  def concat(self, image1, image2, direction, match_image_size):
+    if match_image_size:
+      image2 = torch.nn.functional.interpolate(image2, size=(image1.shape[2], image1.shape[3]), mode="bilinear")
+    if direction == 'right':
+      row = torch.cat((image1, image2), dim=2)
+    elif direction == 'down':
+      row = torch.cat((image1, image2), dim=1)
+    elif direction == 'left':
+      row = torch.cat((image2, image1), dim=2)
+    elif direction == 'up':
+      row = torch.cat((image2, image1), dim=1)
+    return (row,)
+
 # 图片背景移除
 from .briaai.rembg import BriaRMBG, preprocess_image, postprocess_image
 from .libs.utils import get_local_filepath, easySave, install_package
@@ -632,12 +696,12 @@ class imageChooser(PreviewImage):
  def INPUT_TYPES(self):
    return {
      "required":{
-
+        "mode": (['Always Pause', 'Keep Last Selection'], {"default": "Always Pause"}),
      },
      "optional": {
        "images": ("IMAGE",),
      },
-      "hidden": {"prompt": "PROMPT", "my_unique_id": "UNIQUE_ID"},
+      "hidden": {"prompt": "PROMPT", "my_unique_id": "UNIQUE_ID", "extra_pnginfo": "EXTRA_PNGINFO"},
    }

  RETURN_TYPES = ("IMAGE",)
@@ -660,8 +724,7 @@ class imageChooser(PreviewImage):
    else:
      return None

-  def chooser(self, prompt=None, my_unique_id=None, **kwargs):
-
+  def chooser(self, prompt=None, my_unique_id=None, extra_pnginfo=None, **kwargs):
    id = my_unique_id[0]
    if id not in ChooserMessage.stash:
      ChooserMessage.stash[id] = {}
@@ -684,9 +747,19 @@ class imageChooser(PreviewImage):
    images = result['ui']['images']
    PromptServer.instance.send_sync("easyuse-image-choose", {"id": id, "urls": images})

+    # 获取上次选择
+    mode = kwargs.pop('mode', 'Always Pause')
+    last_choosen = None
+    if mode == 'Keep Last Selection':
+      if id and extra_pnginfo[0] and "workflow" in extra_pnginfo[0]:
+        workflow = extra_pnginfo[0]["workflow"]
+        node = next((x for x in workflow["nodes"] if str(x["id"]) == id), None)
+        if node:
+          last_choosen = node['properties']['values']
+
    # wait for selection
    try:
-      selections = ChooserMessage.waitForMessage(id, asList=True)
+      selections = ChooserMessage.waitForMessage(id, asList=True) if last_choosen is None or len(last_choosen)<1 else last_choosen
      choosen = [x for x in selections if x >= 0] if len(selections)>1 else [0]
    except ChooserCancelled:
      raise comfy.model_management.InterruptProcessingException()
@@ -794,6 +867,7 @@ class humanSegmentation:
            "image": ("IMAGE",),
            "method": (["selfie_multiclass_256x256", "human_parsing_lip"],),
            "confidence": ("FLOAT", {"default": 0.4, "min": 0.05, "max": 0.95, "step": 0.01},),
+            "crop_multi": ("FLOAT", {"default": 0.0, "min": 0.0, "max": 10.0, "step": 0.001},),
          },
          "hidden": {
              "prompt": "PROMPT",
@@ -801,8 +875,8 @@ class humanSegmentation:
          }
        }

-    RETURN_TYPES = ("IMAGE", "MASK",)
-    RETURN_NAMES = ("image", "mask",)
+    RETURN_TYPES = ("IMAGE", "MASK", "BBOX")
+    RETURN_NAMES = ("image", "mask", "bbox")
    FUNCTION = "parsing"
    CATEGORY = "EasyUse/Segmentation"

@@ -819,7 +893,7 @@ class humanSegmentation:
        numpy_image = cv2.cvtColor(numpy_image, cv2.COLOR_BGR2RGB)
      return mp.Image(image_format=image_format, data=numpy_image)

-    def parsing(self, image, confidence, method, prompt=None, my_unique_id=None):
+    def parsing(self, image, confidence, method, crop_multi, prompt=None, my_unique_id=None):
      mask_components = []
      if my_unique_id in prompt:
        if prompt[my_unique_id]["inputs"]['mask_components']:
@@ -919,7 +993,253 @@ class humanSegmentation:

        output_image, = JoinImageWithAlpha().join_image_with_alpha(image, alpha)

-      return (output_image, mask)
+      # use crop
+      bbox = [[0, 0, 0, 0]]
+      if crop_multi > 0.0:
+        output_image, mask, bbox = imageCropFromMask().crop(output_image, mask, crop_multi, crop_multi, 1.0)
+
+      return (output_image, mask, bbox)
+
+
+class imageCropFromMask:
+    @classmethod
+    def INPUT_TYPES(s):
+        return {
+          "required": {
+              "image": ("IMAGE",),
+              "mask": ("MASK",),
+              "image_crop_multi": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.001}),
+              "mask_crop_multi": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 10.0, "step": 0.001}),
+              "bbox_smooth_alpha": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0, "step": 0.01}),
+          },
+        }
+
+    RETURN_TYPES = ("IMAGE", "MASK", "BBOX",)
+    RETURN_NAMES = ("crop_image", "crop_mask", "bbox",)
+    FUNCTION = "crop"
+    CATEGORY = "EasyUse/Image"
+
+    def smooth_bbox_size(self, prev_bbox_size, curr_bbox_size, alpha):
+        if alpha == 0:
+            return prev_bbox_size
+        return round(alpha * curr_bbox_size + (1 - alpha) * prev_bbox_size)
+
+    def smooth_center(self, prev_center, curr_center, alpha=0.5):
+        if alpha == 0:
+            return prev_center
+        return (
+            round(alpha * curr_center[0] + (1 - alpha) * prev_center[0]),
+            round(alpha * curr_center[1] + (1 - alpha) * prev_center[1])
+        )
+
+    def image2mask(self, image):
+      return image[:, :, :, 0]
+
+    def mask2image(self, mask):
+      return mask.reshape((-1, 1, mask.shape[-2], mask.shape[-1])).movedim(1, -1).expand(-1, -1, -1, 3)
+
+    def cropimage(self, original_images, masks, crop_size_mult, bbox_smooth_alpha):
+
+      bounding_boxes = []
+      cropped_images = []
+
+      self.max_bbox_width = 0
+      self.max_bbox_height = 0
+
+      # First, calculate the maximum bounding box size across all masks
+      curr_max_bbox_width = 0
+      curr_max_bbox_height = 0
+      for mask in masks:
+        _mask = tensor2pil(mask)
+        non_zero_indices = np.nonzero(np.array(_mask))
+        min_x, max_x = np.min(non_zero_indices[1]), np.max(non_zero_indices[1])
+        min_y, max_y = np.min(non_zero_indices[0]), np.max(non_zero_indices[0])
+        width = max_x - min_x
+        height = max_y - min_y
+        curr_max_bbox_width = max(curr_max_bbox_width, width)
+        curr_max_bbox_height = max(curr_max_bbox_height, height)
+
+      # Smooth the changes in the bounding box size
+      self.max_bbox_width = self.smooth_bbox_size(self.max_bbox_width, curr_max_bbox_width, bbox_smooth_alpha)
+      self.max_bbox_height = self.smooth_bbox_size(self.max_bbox_height, curr_max_bbox_height, bbox_smooth_alpha)
+
+      # Apply the crop size multiplier
+      self.max_bbox_width = round(self.max_bbox_width * crop_size_mult)
+      self.max_bbox_height = round(self.max_bbox_height * crop_size_mult)
+      bbox_aspect_ratio = self.max_bbox_width / self.max_bbox_height
+
+      # Then, for each mask and corresponding image...
+      for i, (mask, img) in enumerate(zip(masks, original_images)):
+        _mask = tensor2pil(mask)
+        non_zero_indices = np.nonzero(np.array(_mask))
+        min_x, max_x = np.min(non_zero_indices[1]), np.max(non_zero_indices[1])
+        min_y, max_y = np.min(non_zero_indices[0]), np.max(non_zero_indices[0])
+
+        # Calculate center of bounding box
+        center_x = np.mean(non_zero_indices[1])
+        center_y = np.mean(non_zero_indices[0])
+        curr_center = (round(center_x), round(center_y))
+
+        # If this is the first frame, initialize prev_center with curr_center
+        if not hasattr(self, 'prev_center'):
+          self.prev_center = curr_center
+
+        # Smooth the changes in the center coordinates from the second frame onwards
+        if i > 0:
+          center = self.smooth_center(self.prev_center, curr_center, bbox_smooth_alpha)
+        else:
+          center = curr_center
+
+        # Update prev_center for the next frame
+        self.prev_center = center
+
+        # Create bounding box using max_bbox_width and max_bbox_height
+        half_box_width = round(self.max_bbox_width / 2)
+        half_box_height = round(self.max_bbox_height / 2)
+        min_x = max(0, center[0] - half_box_width)
+        max_x = min(img.shape[1], center[0] + half_box_width)
+        min_y = max(0, center[1] - half_box_height)
+        max_y = min(img.shape[0], center[1] + half_box_height)
+
+        # Append bounding box coordinates
+        bounding_boxes.append((min_x, min_y, max_x - min_x, max_y - min_y))
+
+        # Crop the image from the bounding box
+        cropped_img = img[min_y:max_y, min_x:max_x, :]
+
+        # Calculate the new dimensions while maintaining the aspect ratio
+        new_height = min(cropped_img.shape[0], self.max_bbox_height)
+        new_width = round(new_height * bbox_aspect_ratio)
+
+        # Resize the image
+        resize_transform = Resize((new_height, new_width))
+        resized_img = resize_transform(cropped_img.permute(2, 0, 1))
+
+        # Perform the center crop to the desired size
+        crop_transform = CenterCrop((self.max_bbox_height, self.max_bbox_width))  # swap the order here if necessary
+        cropped_resized_img = crop_transform(resized_img)
+
+        cropped_images.append(cropped_resized_img.permute(1, 2, 0))
+
+      return cropped_images, bounding_boxes
+
+    def crop(self, image, mask, image_crop_multi, mask_crop_multi, bbox_smooth_alpha):
+      cropped_images, bounding_boxes = self.cropimage(image, mask, image_crop_multi, bbox_smooth_alpha)
+      cropped_mask_image, _ = self.cropimage(self.mask2image(mask), mask, mask_crop_multi, bbox_smooth_alpha)
+
+      cropped_image_out = torch.stack(cropped_images, dim=0)
+      cropped_mask_out = torch.stack(cropped_mask_image, dim=0)
+
+      return (cropped_image_out, cropped_mask_out[:, :, :, 0], bounding_boxes)
+
+
+class imageUncropFromBBOX:
+    @classmethod
+    def INPUT_TYPES(s):
+        return {
+          "required": {
+              "original_image": ("IMAGE",),
+              "crop_image": ("IMAGE",),
+              "bbox": ("BBOX",),
+              "border_blending": ("FLOAT", {"default": 0.25, "min": 0.0, "max": 1.0, "step": 0.01},),
+              "use_square_mask": ("BOOLEAN", {"default": True}),
+          },
+          "optional":{
+            "optional_mask": ("MASK",)
+          }
+        }
+
+    RETURN_TYPES = ("IMAGE",)
+    RETURN_NAMES = ("image",)
+    FUNCTION = "uncrop"
+    CATEGORY = "EasyUse/Image"
+
+    def bbox_check(self, bbox, target_size=None):
+      if not target_size:
+        return bbox
+
+      new_bbox = (
+        bbox[0],
+        bbox[1],
+        min(target_size[0] - bbox[0], bbox[2]),
+        min(target_size[1] - bbox[1], bbox[3]),
+      )
+      return new_bbox
+
+    def bbox_to_region(self, bbox, target_size=None):
+      bbox = self.bbox_check(bbox, target_size)
+      return (bbox[0], bbox[1], bbox[0] + bbox[2], bbox[1] + bbox[3])
+
+    def uncrop(self, original_image, crop_image, bbox, border_blending, use_square_mask, optional_mask=None):
+      def inset_border(image, border_width=20, border_color=(0)):
+        width, height = image.size
+        bordered_image = Image.new(image.mode, (width, height), border_color)
+        bordered_image.paste(image, (0, 0))
+        draw = ImageDraw.Draw(bordered_image)
+        draw.rectangle((0, 0, width - 1, height - 1), outline=border_color, width=border_width)
+        return bordered_image
+
+      if len(original_image) != len(crop_image):
+        raise ValueError(
+          f"The number of original_images ({len(original_image)}) and cropped_images ({len(crop_image)}) should be the same")
+
+        # Ensure there are enough bboxes, but drop the excess if there are more bboxes than images
+      if len(bbox) > len(original_image):
+        print(f"Warning: Dropping excess bounding boxes. Expected {len(original_image)}, but got {len(bbox)}")
+        bbox = bbox[:len(original_image)]
+      elif len(bbox) < len(original_image):
+        raise ValueError("There should be at least as many bboxes as there are original and cropped images")
+
+
+      out_images = []
+
+      for i in range(len(original_image)):
+        img = tensor2pil(original_image[i])
+        crop = tensor2pil(crop_image[i])
+        _bbox = bbox[i]
+
+        bb_x, bb_y, bb_width, bb_height = _bbox
+        paste_region = self.bbox_to_region((bb_x, bb_y, bb_width, bb_height), img.size)
+        
+        # rescale the crop image to fit the paste_region
+        crop = crop.resize((round(paste_region[2] - paste_region[0]), round(paste_region[3] - paste_region[1])))
+        crop_img = crop.convert("RGB")
+
+        # border blending
+        if border_blending > 1.0:
+          border_blending = 1.0
+        elif border_blending < 0.0:
+          border_blending = 0.0
+
+        blend_ratio = (max(crop_img.size) / 2) * float(border_blending)
+        blend = img.convert("RGBA")
+
+        if use_square_mask:
+          mask = Image.new("L", img.size, 0)
+          mask_block = Image.new("L", (paste_region[2] - paste_region[0], paste_region[3] - paste_region[1]), 255)
+          mask_block = inset_border(mask_block, round(blend_ratio / 2), (0))
+          mask.paste(mask_block, paste_region)
+        else:
+          if optional_mask is None:
+            raise ValueError("optional_mask is required when use_square_mask is False")
+          original_mask = tensor2pil(optional_mask)
+          original_mask = original_mask.resize((paste_region[2] - paste_region[0], paste_region[3] - paste_region[1]))
+          mask = Image.new("L", img.size, 0)
+          mask.paste(original_mask, paste_region)
+
+        mask = mask.filter(ImageFilter.BoxBlur(radius=blend_ratio / 4))
+        mask = mask.filter(ImageFilter.GaussianBlur(radius=blend_ratio / 4))
+
+        blend.paste(crop_img, paste_region)
+        blend.putalpha(mask)
+
+        img = Image.alpha_composite(img.convert("RGBA"), blend)
+        out_images.append(img.convert("RGB"))
+
+      output_images = torch.cat([pil2tensor(img) for img in out_images], dim=0)
+      return (output_images,)
+
+

 import cv2
 import base64
@@ -1001,6 +1321,35 @@ class imageToBase64:
      base64_str = base64.b64encode(image_bytes).decode("utf-8")
      return {"result": (base64_str,)}

+class removeLocalImage:
+  @classmethod
+  def INPUT_TYPES(s):
+      return {
+        "required": {
+          "file_name": ("STRING",{"default":""}),
+        },
+      }
+
+  RETURN_TYPES = ()
+  OUTPUT_NODE = True
+  FUNCTION = "remove"
+  CATEGORY = "EasyUse/Image"
+
+  def remove(self, file_name):
+    hasFile = False
+    for file in os.listdir(folder_paths.input_directory):
+      name_without_extension, file_extension = os.path.splitext(file)
+      if name_without_extension == file_name or file == file_name:
+        os.remove(os.path.join(folder_paths.input_directory, file))
+        hasFile = True
+        break
+    if hasFile:
+      PromptServer.instance.send_sync("easyuse-toast", {"content": "Removed SuccessFully", "type":'success'})
+    else:
+      PromptServer.instance.send_sync("easyuse-toast", {"content": "Removed Failed", "type": 'error'})
+    return ()
+
+
 # 姿势编辑器
 class poseEditor:
  @classmethod
@@ -1055,8 +1404,12 @@ NODE_CLASS_MAPPINGS = {
  "easy imageScaleDownToSize": imageScaleDownToSize,
  "easy imageRatio": imageRatio,
  "easy imageToMask": imageToMask,
+  "easy imageConcat": imageConcat,
  "easy imageSplitList": imageSplitList,
+  "easy imageSplitGrid": imageSplitGrid,
  "easy imagesSplitImage": imagesSplitImage,
+  "easy imageCropFromMask": imageCropFromMask,
+  "easy imageUncropFromBBOX": imageUncropFromBBOX,
  "easy imageSave": imageSaveSimple,
  "easy imageRemBg": imageRemBg,
  "easy imageChooser": imageChooser,
@@ -1066,6 +1419,7 @@ NODE_CLASS_MAPPINGS = {
  "easy imageToBase64": imageToBase64,
  "easy joinImageBatch": JoinImageBatch,
  "easy humanSegmentation": humanSegmentation,
+  "easy removeLocalImage": removeLocalImage,
  "easy poseEditor": poseEditor
 }

@@ -1082,8 +1436,12 @@ NODE_DISPLAY_NAME_MAPPINGS = {
  "easy imageRatio": "ImageRatio",
  "easy imageToMask": "ImageToMask",
  "easy imageHSVMask": "ImageHSVMask",
+  "easy imageConcat": "imageConcat",
  "easy imageSplitList": "imageSplitList",
+  "easy imageSplitGrid": "imageSplitGrid",
  "easy imagesSplitImage": "imagesSplitImage",
+  "easy imageCropFromMask": "imageCropFromMask",
+  "easy imageUncropFromBBOX": "imageUncropFromBBOX",
  "easy imageSave": "SaveImage (Simple)",
  "easy imageRemBg": "Image Remove Bg",
  "easy imageChooser": "Image Chooser",
@@ -1093,5 +1451,6 @@ NODE_DISPLAY_NAME_MAPPINGS = {
  "easy loadImageBase64": "Load Image (Base64)",
  "easy imageToBase64": "Image To Base64",
  "easy humanSegmentation": "Human Segmentation",
+  "easy removeLocalImage": "Remove Local Image",
  "easy poseEditor": "PoseEditor",
 }
@@ -17,7 +17,7 @@ try:
    from diffusers.models.modeling_utils import ModelMixin
    from diffusers import __version__
    if __version__:
-        if version.parse(__version__) < version.parse("0.27.0"):
+        if version.parse(__version__) < version.parse("0.26.0"):
            from diffusers.models.unet_2d_blocks import UNetMidBlock2D, get_down_block, get_up_block
        else:
            from diffusers.models.unets.unet_2d_blocks import UNetMidBlock2D, get_down_block, get_up_block
@@ -7,26 +7,12 @@ class easyControlnet:
    def __init__(self):
        pass

-    def load_controlnet(self, control_net_name, control_net, scale_soft_weights):
-        if control_net is None:
-            if scale_soft_weights < 1:
-                if "ScaledSoftControlNetWeights" in NODE_CLASS_MAPPINGS:
-                    soft_weight_cls = NODE_CLASS_MAPPINGS['ScaledSoftControlNetWeights']
-                    (weights, timestep_keyframe) = soft_weight_cls().load_weights(scale_soft_weights, False)
-                    cn_adv_cls = NODE_CLASS_MAPPINGS['ControlNetLoaderAdvanced']
-                    control_net, = cn_adv_cls().load_controlnet(control_net_name, timestep_keyframe)
-                else:
-                    raise Exception(f"[Advanced-ControlNet Not Found] you need to install 'COMFYUI-Advanced-ControlNet'")
-            else:
-                controlnet_path = folder_paths.get_full_path("controlnet", control_net_name)
-                control_net = comfy.controlnet.load_controlnet(controlnet_path)
-        return control_net
-
-    def apply(self, control_net_name, image, positive, negative, strength, start_percent=0, end_percent=1, control_net=None, scale_soft_weights=1, mask=None):
+    def apply(self, control_net_name, image, positive, negative, strength, start_percent=0, end_percent=1, control_net=None, scale_soft_weights=1, mask=None, easyCache=None):
        if strength == 0:
            return (positive, negative)

-        control_net = self.load_controlnet(control_net_name, control_net, scale_soft_weights)
+        if control_net is None:
+            control_net = easyCache.load_controlnet(control_net_name, scale_soft_weights)

        if mask is not None:
            mask = mask.to(self.device)
@@ -5,6 +5,7 @@ import numpy as np
 from enum import Enum
 from PIL import Image
 from io import BytesIO
+from typing import List, Union

 import folder_paths
 from .utils import install_package
@@ -15,6 +16,17 @@ def pil2tensor(image):
 # Tensor to PIL
 def tensor2pil(image):
    return Image.fromarray(np.clip(255. * image.cpu().numpy().squeeze(), 0, 255).astype(np.uint8))
+# np to Tensor
+def np2tensor(img_np: Union[np.ndarray, List[np.ndarray]]) -> torch.Tensor:
+  if isinstance(img_np, list):
+    return torch.cat([np2tensor(img) for img in img_np], dim=0)
+  return torch.from_numpy(img_np.astype(np.float32) / 255.0).unsqueeze(0)
+# Tensor to np
+def tensor2np(tensor: torch.Tensor) -> List[np.ndarray]:
+  if len(tensor.shape) == 3:  # Single image
+    return np.clip(255.0 * tensor.cpu().numpy(), 0, 255).astype(np.uint8)
+  else:  # Batch of images
+    return [np.clip(255.0 * t.cpu().numpy(), 0, 255).astype(np.uint8) for t in tensor]

 def pil2byte(pil_image, format='PNG'):
  byte_arr = BytesIO()
@@ -48,6 +60,14 @@ def image2mask(image: Image) -> torch.Tensor:
  ret_mask = torch.tensor([pil2tensor(_image)[0, :, :, 3].tolist()])
  return ret_mask

+def mask2image(mask: torch.Tensor) -> Image:
+  masks = tensor2np(mask)
+  for m in masks:
+    _mask = Image.fromarray(m).convert("L")
+    _image = Image.new("RGBA", _mask.size, color='white')
+    _image = Image.composite(
+      _image, Image.new("RGBA", _mask.size, color='black'), _mask)
+  return _image

 # 图像融合
 class blendImage:
@@ -1,16 +1,19 @@
 import time, os, psutil
 import comfy.utils
 import comfy.sd
+import comfy.controlnet
 import folder_paths
 from nodes import NODE_CLASS_MAPPINGS
 from collections import defaultdict
 from ..log import log_node_info, log_node_error

-stable_diffusion_loaders = ["easy a1111Loader", "easy comfyLoader", "easy zero123Loader", "easy svdLoader"]
+stable_diffusion_loaders = ["easy fullLoader", "easy a1111Loader", "easy comfyLoader", "easy zero123Loader", "easy svdLoader"]
 stable_cascade_loaders = ["easy cascadeLoader"]
+controlnet_loaders = ["easy controlnetLoader", "easy controlnetLoaderADV"]
+instant_loaders = ["easy instantIDApply", "easy instantIDApplyADV"]
 cascade_vae_node = ["easy preSamplingCascade", "easy fullCascadeKSampler"]
 model_merge_node = ["easy XYInputs: ModelMergeBlocks"]
-lora_widget = ["easy a1111Loader", "easy comfyLoader"]
+lora_widget = ["easy fullLoader", "easy a1111Loader", "easy comfyLoader"]

 class easyLoader:
    def __init__(self):
@@ -22,8 +25,9 @@ class easyLoader:
            "bvae": defaultdict(tuple),
            "vae": defaultdict(object),
            "lora": defaultdict(dict),  # {lora_name: {UID: (model_lora, clip_lora)}}
+            "controlnet": defaultdict(dict),
        }
-        self.memory_threshold = self.determine_memory_threshold(0.7)
+        self.memory_threshold = self.determine_memory_threshold(0.9)
        self.lora_name_cache = []

    def clean_values(self, values: str):
@@ -50,9 +54,17 @@ class easyLoader:
        for key in keys - desired_names:
            del self.loaded_objects[object_type][key]

-    def get_input_value(self, entry, key):
+    def get_input_value(self, entry, key, prompt=None):
        val = entry["inputs"][key]
-        return val if isinstance(val, str) else val[0]
+        if isinstance(val, str):
+            return val
+        elif isinstance(val, list):
+            if prompt is not None and val[0]:
+                return prompt[val[0]]['inputs'][key]
+            else:
+                return val[0]
+        else:
+            return str(val)

    def process_pipe_loader(self, entry, desired_ckpt_names, desired_vae_names, desired_lora_names, desired_lora_settings, num_loras=3, suffix=""):
        for idx in range(1, num_loras + 1):
@@ -71,10 +83,10 @@ class easyLoader:
        desired_vae_names = set()
        desired_lora_names = set()
        desired_lora_settings = set()
+        desired_controlnet_names = set()

        for entry in prompt.values():
            class_type = entry["class_type"]
-
            if class_type in lora_widget:
                lora_name = self.get_input_value(entry, "lora_name")
                desired_lora_names.add(lora_name)
@@ -82,7 +94,7 @@ class easyLoader:
                desired_lora_settings.add(setting)

            if class_type in stable_diffusion_loaders:
-                desired_ckpt_names.add(self.get_input_value(entry, "ckpt_name"))
+                desired_ckpt_names.add(self.get_input_value(entry, "ckpt_name", prompt))
                desired_vae_names.add(self.get_input_value(entry, "vae_name"))

            elif class_type in stable_cascade_loaders:
@@ -99,6 +111,16 @@ class easyLoader:
                if decode_vae_name and decode_vae_name != 'None':
                    desired_vae_names.add(decode_vae_name)

+            elif class_type in controlnet_loaders:
+                control_net_name = self.get_input_value(entry, "control_net_name", prompt)
+                scale_soft_weights = self.get_input_value(entry, "scale_soft_weights")
+                desired_controlnet_names.add(f'{control_net_name};{scale_soft_weights}')
+
+            elif class_type in instant_loaders:
+                control_net_name = self.get_input_value(entry, "control_net_name", prompt)
+                scale_soft_weights = self.get_input_value(entry, "cn_soft_weights")
+                desired_controlnet_names.add(f'{control_net_name};{scale_soft_weights}')
+
            elif class_type in model_merge_node:
                desired_ckpt_names.add(self.get_input_value(entry, "ckpt_name_1"))
                desired_ckpt_names.add(self.get_input_value(entry, "ckpt_name_2"))
@@ -106,7 +128,7 @@ class easyLoader:
                if vae_use != 'Use Model 1' and vae_use != 'Use Model 2':
                    desired_vae_names.add(vae_use)

-        object_types = ["ckpt", "unet", "clip", "bvae", "vae", "lora"]
+        object_types = ["ckpt", "unet", "clip", "bvae", "vae", "lora", "controlnet"]
        for object_type in object_types:
            if object_type == 'unet':
                desired_names = desired_unet_names
@@ -117,6 +139,8 @@ class easyLoader:
                    desired_names = desired_ckpt_names
            elif object_type == "vae":
                desired_names = desired_vae_names
+            elif object_type == "controlnet":
+                desired_names = desired_controlnet_names
            else:
                desired_names = desired_lora_names
            self.clear_unused_objects(desired_names, object_type)
@@ -155,7 +179,7 @@ class easyLoader:
        current_memory = self.get_memory_usage()
        if current_memory < self.memory_threshold:
            return
-        eviction_order = ["vae", "lora", "bvae", "clip", "ckpt"]
+        eviction_order = ["vae", "lora", "bvae", "clip", "ckpt", "controlnet"]
        for obj_type in eviction_order:
            if current_memory < self.memory_threshold:
                break
@@ -225,6 +249,25 @@ class easyLoader:

        return model

+    def load_controlnet(self, control_net_name, scale_soft_weights=1):
+        unique_id = f'{control_net_name};{str(scale_soft_weights)}'
+        if unique_id in self.loaded_objects["controlnet"]:
+            return self.loaded_objects["controlnet"][unique_id][0]
+        if scale_soft_weights < 1:
+            if "ScaledSoftControlNetWeights" in NODE_CLASS_MAPPINGS:
+                soft_weight_cls = NODE_CLASS_MAPPINGS['ScaledSoftControlNetWeights']
+                (weights, timestep_keyframe) = soft_weight_cls().load_weights(scale_soft_weights, False)
+                cn_adv_cls = NODE_CLASS_MAPPINGS['ControlNetLoaderAdvanced']
+                control_net, = cn_adv_cls().load_controlnet(control_net_name, timestep_keyframe)
+            else:
+                raise Exception(
+                    f"[Advanced-ControlNet Not Found] you need to install 'COMFYUI-Advanced-ControlNet'")
+        else:
+            controlnet_path = folder_paths.get_full_path("controlnet", control_net_name)
+            control_net = comfy.controlnet.load_controlnet(controlnet_path)
+        self.add_to_cache("controlnet", unique_id, control_net)
+        self.eviction_based_on_memory()
+        return control_net
    def load_clip(self, clip_name, type='stable_diffusion'):
        if type == 'stable_diffusion':
            clip_type = comfy.sd.CLIPType.STABLE_DIFFUSION
@@ -47,6 +47,14 @@ class easySampler:
            parts.append('None')
        return parts

+    def add_model_patch_option(self, model):
+        if 'transformer_options' not in model.model_options:
+            model.model_options['transformer_options'] = {}
+        to = model.model_options['transformer_options']
+        if "model_patch" not in to:
+            to["model_patch"] = {}
+        return to
+
    def common_ksampler(self, model, seed, steps, cfg, sampler_name, scheduler, positive, negative, latent, denoise=1.0,
                        disable_noise=False, start_step=None, last_step=None, force_full_denoise=False,
                        preview_latent=True, disable_pbar=False, custom=None):
@@ -91,6 +99,26 @@ class easySampler:
                batch_inds = latent["batch_index"] if "batch_index" in latent else None
                noise = comfy.sample.prepare_noise(latent_image, seed, batch_inds)

+            #######################################################################################
+            # brushnet
+            transformer_options = model.model_options['transformer_options'] if "transformer_options" in model.model_options else {}
+            if 'model_patch' in transformer_options and 'brushnet' in transformer_options['model_patch']:
+                to = self.add_model_patch_option(model)
+                mp = to['model_patch']
+                if isinstance(model.model.model_config, comfy.supported_models.SD15):
+                    mp['SDXL'] = False
+                elif isinstance(model.model.model_config, comfy.supported_models.SDXL):
+                    mp['SDXL'] = True
+                else:
+                    print('Base model type: ', type(model.model.model_config))
+                    raise Exception("Unsupported model type: ", type(model.model.model_config))
+
+                mp['unet'] = model.model.diffusion_model
+                mp['step'] = 0
+                mp['total_steps'] = 1
+
+            #
+            #######################################################################################
            samples = comfy.sample.sample(model, noise, steps, cfg, sampler_name, scheduler, positive, negative,
                                          latent_image,
                                          denoise=denoise, disable_noise=disable_noise, start_step=start_step,
@@ -129,8 +157,31 @@ class easySampler:

        pbar = comfy.utils.ProgressBar(steps)

+        #######################################################################################
+        # brushnet
+        to = None
+        transformer_options = model.model_options['transformer_options'] if "transformer_options" in model.model_options else {}
+        if 'model_patch' in transformer_options and 'brushnet_model' in transformer_options['model_patch']:
+            to = self.add_model_patch_option(model)
+            mp = to['model_patch']
+            if isinstance(model.model.model_config, comfy.supported_models.SD15):
+                mp['SDXL'] = False
+            elif isinstance(model.model.model_config, comfy.supported_models.SDXL):
+                mp['SDXL'] = True
+            else:
+                print('Base model type: ', type(model.model.model_config))
+                raise Exception("Unsupported model type: ", type(model.model.model_config))
+
+            mp['unet'] = model.model.diffusion_model
+            mp['step'] = 0
+            mp['total_steps'] = 1
+        #
+        #######################################################################################
+
        def callback(step, x0, x, total_steps):
            preview_bytes = None
+            if to is not None and "model_patch" in to:
+                to['model_patch']['step'] = step + 1
            if previewer:
                preview_bytes = previewer.decode_latent_to_preview_image(preview_format, x0)
            pbar.update_absolute(step + 1, total_steps, preview_bytes)
@@ -21,6 +21,8 @@ def get_comfyui_revision():
 import sys
 import importlib.util
 import importlib.metadata
+import comfy.model_management as mm
+import gc
 from packaging import version
 from server import PromptServer
 def is_package_installed(package):
@@ -81,16 +83,6 @@ def find_tags(string: str, sep="/") -> list[str]:
        return string.split(sep)[:-1]
    return []

-import folder_paths
-def add_folder_path_and_extensions(folder_name, full_folder_paths, extensions):
-    for full_folder_path in full_folder_paths:
-        folder_paths.add_model_folder_path(folder_name, full_folder_path)
-    if folder_name in folder_paths.folder_names_and_paths:
-        current_paths, current_extensions = folder_paths.folder_names_and_paths[folder_name]
-        updated_extensions = current_extensions | extensions
-        folder_paths.folder_names_and_paths[folder_name] = (current_paths, updated_extensions)
-    else:
-        folder_paths.folder_names_and_paths[folder_name] = (full_folder_paths, extensions)

 from comfy.model_base import BaseModel
 import comfy.supported_models
@@ -259,8 +251,6 @@ def getMetadata(filepath):
        return header

 def cleanGPUUsedForce():
-    import torch.cuda
-    import comfy.model_management
-    if torch.cuda.is_available():
-        torch.cuda.empty_cache()
-    comfy.model_management.unload_all_models()
+    gc.collect()
+    mm.unload_all_models()
+    mm.soft_empty_cache()
@@ -0,0 +1,15 @@
+[project]
+name = "comfyui-easy-use"
+description = "To enhance the usability of ComfyUI, optimizations and integrations have been implemented for several commonly used nodes."
+version = "1.1.7"
+license = "LICENSE"
+dependencies = ["diffusers>=0.25.0", "clip_interrogator>=0.6.0", "onnxruntime", "aiohttp"]
+
+[project.urls]
+Repository = "https://github.com/yolain/ComfyUI-Easy-Use"
+#  Used by Comfy Registry https://comfyregistry.org
+
+[tool.comfy]
+PublisherId = "yolain"
+DisplayName = "ComfyUI-Easy-Use"
+Icon = ""
@@ -27,4 +27,9 @@
 }
 .easyuse-chooser-dialog-images img.selected{
    border: 4px solid var(--success-color);
+}
+
+.easyuse-chooser-hidden{
+    display: none;
+    height:0;
 }
@@ -10,4 +10,23 @@
  background-color: var(--comfy-menu-bg);
  padding: 10px 4px;
  border: 1px solid var(--border-color);z-index: 999999999;padding-top: 0;
+}
+#easyuse_groups_map .icon{
+  width: 12px;
+  height:12px;
+}
+#easyuse_groups_map .closeBtn{
+  float: right;
+  color: var(--input-text);
+  border-radius:30px;
+  background-color: var(--comfy-input-bg);
+  border: 1px solid var(--border-color);
+  cursor: pointer;
+  aspect-ratio: 1 / 1;
+  display: flex;
+  justify-content: center;
+  align-items: center;
+}
+#easyuse_groups_map .closeBtn:hover{
+  filter:brightness(120%);
 }
@@ -1,7 +1,7 @@
 .easyuse-toolbar{
-    background: rgba(35,35,35,.5);
+    background: rgba(15,15,15,.5);
    backdrop-filter: blur(4px) brightness(120%);
-    border-radius:0 15px 15px 0;
+    border-radius:0 12px 12px 0;
    min-width:50px;
    height:24px;
    position: fixed;
@@ -67,18 +67,63 @@


 .easyuse-guide-dialog{
-    min-width: 600px;
+    max-width: 300px;
    font-family: var(--font-family);
+    position: absolute;
+    z-index:100;
+    left:0;
+    bottom:140px;
+    background: rgba(25,25,25,.85);
+    backdrop-filter: blur(8px) brightness(120%);
+    border-radius:0 12px 12px 0;
+    padding:10px;
+    transition: .5s all ease-in-out;
+    visibility: visible;
+    opacity: 1;
+    transform: translateX(0%);
+}
+.easyuse-guide-dialog.disable-render-info{
+    bottom:110px;
+}
+.easyuse-guide-dialog-top{
+    display: flex;
+    justify-content: space-between;
+    align-items: center;
+}
+.easyuse-guide-dialog-top .icon{
+    width: 12px;
+    height:12px;
+}
+.easyuse-guide-dialog.hidden{
+    opacity: 0;
+    transform: translateX(-50%);
+    visibility: hidden;
+}
+.easyuse-guide-dialog .closeBtn{
+  float: right;
+  color: var(--input-text);
+  border-radius:30px;
+  background-color: var(--comfy-input-bg);
+  border: 1px solid var(--border-color);
+  cursor: pointer;
+  aspect-ratio: 1 / 1;
+  display: flex;
+  justify-content: center;
+  align-items: center;
+}
+.easyuse-guide-dialog .closeBtn:hover{
+    filter:brightness(120%);
 }
 .easyuse-guide-dialog-title{
    color:var(--input-text);
-    font-size: 20px;
+    font-size: 16px;
    font-weight: bold;
    margin-bottom: 5px;
 }
 .easyuse-guide-dialog-remark{
    color: var(--input-text);
-    font-size: 14px;
+    font-size: 12px;
+    margin-top: 5px;
 }
 .easyuse-guide-dialog-content{
    max-height: 600px;
@@ -91,7 +136,6 @@
 .easyuse-guide-dialog-note{
    margin-top: 20px;
    color:white;
-    font-size: 14px;
 }
 .easyuse-guide-dialog p{
    margin:4px 0;
@@ -99,19 +143,31 @@
    font-weight: 300;
 }
 .markdown-body h1, .markdown-body h2, .markdown-body h3, .markdown-body h4, .markdown-body h5, .markdown-body h6 {
-    margin-top: 24px;
-    margin-bottom: 16px;
+    margin-top: 12px;
+    margin-bottom: 8px;
    font-weight: 600;
    line-height: 1.25;
-    padding-bottom: 10px;
+    padding-bottom: 5px;
    border-bottom: 1px solid var(--border-color);
    color: var(--input-text);
 }
+.markdown-body h1{
+    font-size: 18px;
+}
+.markdown-body h2{
+    font-size: 16px;
+}
+.markdown-body h3{
+    font-size: 14px;
+}
+.markdown-body h4{
+    font-size: 13px;
+}
 .markdown-body table {
    display: block;
-    width: 100%;
-    width: max-content;
-    max-width: 100%;
+    /*width: 100%;*/
+    /*width: max-content;*/
+    max-width: 300px;
    overflow: auto;
    color:var(--input-text);
    box-sizing: border-box;
@@ -121,7 +177,7 @@
 }
 .markdown-body table th, .markdown-body table td {
    padding: 6px 13px;
-    font-size: 14px;
+    font-size: 12px;
    margin:0;
    border-right: 1px solid var(--border-color);
    border-bottom: 1px solid var(--border-color);
@@ -138,14 +194,11 @@
 .markdown-body table th{
    font-weight: bold;
    width: auto;
-    min-width: 100px;
+    min-width: 70px;
 }
 .markdown-body table th:last-child{
    width:100%;
 }
-.markdown-body a{
-    margin-right: 10px;
-}
 .markdown-body .warning{
    color:var(--warning-color)
 }
@@ -26,6 +26,7 @@ const zhCN = {
    "Trained Words": "训练词",
    "BaseModel": "基础算法",
    "Details": "详情",
+    "Description": "描述",
    "Download": "下载量",
    "Source": "来源",
    "Saving Preview...": "正在保存预览图...",
@@ -1,4 +1,5 @@
 export const quesitonIcon = `<svg t="1714564780771" class="icon" viewBox="0 0 1024 1024" version="1.1" xmlns="http://www.w3.org/2000/svg" p-id="1489" width="200" height="200" data-spm-anchor-id="a313x.search_index.0.i2.5a663a81pw6qup"><path d="M514.048 54.272q95.232 0 178.688 36.352t145.92 98.304 98.304 145.408 35.84 178.688-35.84 178.176-98.304 145.408-145.92 98.304-178.688 35.84-178.176-35.84-145.408-98.304-98.304-145.408-35.84-178.176 35.84-178.688 98.304-145.408 145.408-98.304 178.176-36.352zM515.072 826.368q26.624 0 44.544-17.92t17.92-43.52q0-26.624-17.92-44.544t-44.544-17.92-44.544 17.92-17.92 44.544q0 25.6 17.92 43.52t44.544 17.92zM567.296 574.464q-1.024-16.384 20.48-34.816t48.128-40.96 49.152-50.688 24.576-65.024q2.048-39.936-8.192-74.752t-33.792-59.904-60.928-39.936-87.552-14.848q-62.464 0-103.936 22.016t-67.072 53.248-35.84 64.512-9.216 55.808q1.024 26.624 16.896 38.912t34.304 12.8 33.792-10.24 15.36-31.232q0-12.288 7.68-30.208t20.992-34.304 32.256-27.648 42.496-11.264q46.08 0 73.728 23.04t25.6 57.856q0 17.408-10.24 32.256t-26.112 28.672-33.792 27.648-33.792 28.672-26.624 32.256-11.776 37.888l1.024 38.912q0 15.36 14.336 29.184t37.888 14.848q23.552-1.024 37.376-15.36t12.8-32.768l0-24.576z" p-id="1490" fill="currentColor"></path></svg>`
 export const rocketIcon = `<svg t="1714565020764" class="icon" viewBox="0 0 1024 1024" version="1.1" xmlns="http://www.w3.org/2000/svg" p-id="7999" width="200" height="200"><path d="M810.438503 379.664884l-71.187166-12.777183C737.426025 180.705882 542.117647 14.602496 532.991087 7.301248c-12.777184-10.951872-32.855615-10.951872-47.45811 0-9.12656 7.301248-204.434938 175.229947-206.26025 359.586453l-67.536542 10.951871c-18.253119 3.650624-31.030303 18.253119-31.030303 36.506239v189.832442c0 10.951872 5.475936 21.903743 12.777184 27.379679 7.301248 5.475936 14.602496 9.12656 23.729055 9.12656h5.475936l133.247772-23.729055c40.156863 47.458111 91.265597 73.012478 151.500891 73.012477 60.235294 0 111.344029-27.379679 151.500891-74.837789l136.898396 23.729055h5.475936c9.12656 0 16.427807-3.650624 23.729055-9.12656 9.12656-7.301248 12.777184-16.427807 12.777184-27.379679V412.520499c1.825312-14.602496-10.951872-29.204991-27.379679-32.855615zM620.606061 766.631016H401.568627c-20.078431 0-36.506239 16.427807-36.506238 36.506239v109.518716c0 14.602496 9.12656 29.204991 23.729055 34.680927 14.602496 5.475936 31.030303 1.825312 40.156863-9.126559l16.427807-18.25312 32.855615 80.313726c5.475936 14.602496 18.253119 23.729055 34.680927 23.729055 16.427807 0 27.379679-9.12656 34.680927-23.729055l32.855615-80.313726 16.427807 18.25312c10.951872 10.951872 25.554367 14.602496 40.156863 9.126559 14.602496-5.475936 23.729055-18.253119 23.729055-34.680927v-109.518716c-3.650624-20.078431-20.078431-36.506239-40.156862-36.506239z" fill="currentColor" p-id="8000"></path></svg>`
 export const groupIcon = `<svg t="1714565543756" class="icon" viewBox="0 0 1024 1024" version="1.1" xmlns="http://www.w3.org/2000/svg" p-id="22538" width="200" height="200"><path d="M871.616 64H152.384c-31.488 0-60.416 25.28-60.416 58.24v779.52c0 32.896 26.24 58.24 60.352 58.24h719.232c34.112 0 60.352-25.344 60.352-58.24V122.24c0.128-32.96-28.8-58.24-60.288-58.24zM286.272 512c-23.616 0-44.672-20.224-44.672-43.008 0-22.784 20.992-43.008 44.608-43.008 23.616 0 44.608 20.224 44.608 43.008A43.328 43.328 0 0 1 286.272 512z m0-202.496c-23.616 0-44.608-20.224-44.608-43.008 0-22.784 20.992-43.008 44.608-43.008 23.616 0 44.608 20.224 44.608 43.008a43.456 43.456 0 0 1-44.608 43.008zM737.728 512H435.904c-23.68 0-44.672-20.224-44.672-43.008 0-22.784 20.992-43.008 44.608-43.008h299.264c23.616 0 44.608 20.224 44.608 43.008a42.752 42.752 0 0 1-41.984 43.008z m0-202.496H435.904c-23.616 0-44.608-20.224-44.608-43.008 0-22.784 20.992-43.008 44.608-43.008h299.264c23.616 0 44.608 20.224 44.608 43.008a42.88 42.88 0 0 1-42.048 43.008z" p-id="22539" fill="currentColor"></path></svg>`
-export const rebootIcon = `<svg t="1714568501931" class="icon" viewBox="0 0 1024 1024" version="1.1" xmlns="http://www.w3.org/2000/svg" p-id="4275" width="200" height="200"><path d="M511.721751 0.000278a511.999861 511.999861 0 1 0 512.277971 511.721751A511.721751 511.721751 0 0 0 511.721751 0.000278zM184.386696 511.722029A36.988583 36.988583 0 0 1 222.487718 475.011556h92.888622a36.710473 36.710473 0 0 1 0 73.420947H222.487718a36.710473 36.710473 0 0 1-38.101022-36.710474z m201.351385 158.522499l-65.911986 65.911987a36.988583 36.988583 0 0 1-62.852781-25.864197 38.101022 38.101022 0 0 1 10.846276-26.142307L333.731577 618.238024a36.710473 36.710473 0 1 1 52.006504 52.006504z m29.201513-256.138985a36.710473 36.710473 0 0 1-52.006504 0l-65.633877-65.633877a36.988583 36.988583 0 0 1 26.142307-62.85278 36.154254 36.154254 0 0 1 25.864197 10.846276L414.939594 361.54282a36.988583 36.988583 0 0 1 0 52.562723z m135.439398 373.779366a37.266693 37.266693 0 0 1-36.988583 36.988583 36.988583 36.988583 0 0 1-36.710473-36.988583V695.274397a36.988583 36.988583 0 0 1 36.710473-36.988583A37.266693 37.266693 0 0 1 550.378992 695.274397z m0-459.437137a37.266693 37.266693 0 0 1-36.988583 36.988583 36.988583 36.988583 0 0 1-36.710473-36.988583V235.559149a36.988583 36.988583 0 0 1 36.710473-36.988583 37.544802 37.544802 0 0 1 36.988583 36.988583z m63.965219 15.85225L679.978088 278.109926a36.710473 36.710473 0 0 1 52.006504 51.728394L667.463154 396.584635a37.544802 37.544802 0 0 1-52.284614 0 36.988583 36.988583 0 0 1-10.568166-26.142306 36.432364 36.432364 0 0 1 9.733837-26.142307z m122.090135 397.974905a37.544802 37.544802 0 0 1-52.284613 0l-65.355767-65.911986a36.154254 36.154254 0 0 1 0-51.728395 36.710473 36.710473 0 0 1 25.864197-10.846276 35.876145 35.876145 0 0 1 25.864197 10.846276l65.911986 65.633877a36.988583 36.988583 0 0 1 0 52.006504z m66.468206-194.676753h-92.888622a36.710473 36.710473 0 0 1 0-73.420947h92.888622a36.710473 36.710473 0 0 1 0 73.420947z" fill="currentColor" p-id="4276"></path></svg>`
+export const rebootIcon = `<svg t="1714568501931" class="icon" viewBox="0 0 1024 1024" version="1.1" xmlns="http://www.w3.org/2000/svg" p-id="4275" width="200" height="200"><path d="M511.721751 0.000278a511.999861 511.999861 0 1 0 512.277971 511.721751A511.721751 511.721751 0 0 0 511.721751 0.000278zM184.386696 511.722029A36.988583 36.988583 0 0 1 222.487718 475.011556h92.888622a36.710473 36.710473 0 0 1 0 73.420947H222.487718a36.710473 36.710473 0 0 1-38.101022-36.710474z m201.351385 158.522499l-65.911986 65.911987a36.988583 36.988583 0 0 1-62.852781-25.864197 38.101022 38.101022 0 0 1 10.846276-26.142307L333.731577 618.238024a36.710473 36.710473 0 1 1 52.006504 52.006504z m29.201513-256.138985a36.710473 36.710473 0 0 1-52.006504 0l-65.633877-65.633877a36.988583 36.988583 0 0 1 26.142307-62.85278 36.154254 36.154254 0 0 1 25.864197 10.846276L414.939594 361.54282a36.988583 36.988583 0 0 1 0 52.562723z m135.439398 373.779366a37.266693 37.266693 0 0 1-36.988583 36.988583 36.988583 36.988583 0 0 1-36.710473-36.988583V695.274397a36.988583 36.988583 0 0 1 36.710473-36.988583A37.266693 37.266693 0 0 1 550.378992 695.274397z m0-459.437137a37.266693 37.266693 0 0 1-36.988583 36.988583 36.988583 36.988583 0 0 1-36.710473-36.988583V235.559149a36.988583 36.988583 0 0 1 36.710473-36.988583 37.544802 37.544802 0 0 1 36.988583 36.988583z m63.965219 15.85225L679.978088 278.109926a36.710473 36.710473 0 0 1 52.006504 51.728394L667.463154 396.584635a37.544802 37.544802 0 0 1-52.284614 0 36.988583 36.988583 0 0 1-10.568166-26.142306 36.432364 36.432364 0 0 1 9.733837-26.142307z m122.090135 397.974905a37.544802 37.544802 0 0 1-52.284613 0l-65.355767-65.911986a36.154254 36.154254 0 0 1 0-51.728395 36.710473 36.710473 0 0 1 25.864197-10.846276 35.876145 35.876145 0 0 1 25.864197 10.846276l65.911986 65.633877a36.988583 36.988583 0 0 1 0 52.006504z m66.468206-194.676753h-92.888622a36.710473 36.710473 0 0 1 0-73.420947h92.888622a36.710473 36.710473 0 0 1 0 73.420947z" fill="currentColor" p-id="4276"></path></svg>`
+export const closeIcon = `<svg t="1714965640187" class="icon" viewBox="0 0 1024 1024" version="1.1" xmlns="http://www.w3.org/2000/svg" p-id="4264" width="200" height="200"><path d="M597.795527 511.488347 813.564755 295.718095c23.833825-23.833825 23.833825-62.47489 0.001023-86.307691-23.832801-23.832801-62.47489-23.833825-86.307691 0L511.487835 425.180656 295.717583 209.410404c-23.833825-23.833825-62.475913-23.833825-86.307691 0-23.832801 23.832801-23.833825 62.47489 0 86.308715l215.769228 215.769228L209.410915 727.258599c-23.833825 23.833825-23.833825 62.47489 0 86.307691 23.832801 23.833825 62.473867 23.833825 86.307691 0l215.768205-215.768205 215.769228 215.769228c23.834848 23.833825 62.475913 23.832801 86.308715 0 23.833825-23.833825 23.833825-62.47489 0-86.307691L597.795527 511.488347z" fill="currentColor" p-id="4265"></path></svg>`
@@ -1,7 +1,7 @@
 import { api } from "../../../../scripts/api.js";
 import { app } from "../../../../scripts/app.js";
 import {deepEqual, addCss, isLocalNetwork} from "../common/utils.js";
-import {quesitonIcon, rocketIcon, groupIcon, rebootIcon} from "../common/icon.js";
+import {quesitonIcon, rocketIcon, groupIcon, rebootIcon, closeIcon} from "../common/icon.js";
 import {$t} from '../common/i18n.js';
 import {toast} from "../common/toast.js";
 import {$el, ComfyDialog} from "../../../../scripts/ui.js";
@@ -29,12 +29,15 @@ api.addEventListener("easyuse-toast",event=>{
 let draggerEl = null
 let isGroupMapcanMove = true
 function createGroupMap(){
+    let div = document.querySelector('#easyuse_groups_map')
+    if(div){
+        div.style.display = div.style.display == 'none' ? 'flex' : 'none'
+        return
+    }
    let groups = app.canvas.graph._groups
    let nodes = app.canvas.graph._nodes
    let old_nodes = groups.length
-    let div =
-        document.querySelector('#easyuse_groups_map') ||
-        document.createElement('div')
+    div = document.createElement('div')
    div.id = 'easyuse_groups_map'
    div.innerHTML = ''
    let btn = document.createElement('div')
@@ -44,18 +47,15 @@ function createGroupMap(){
        align-items: center;
        padding: 0 6px;
        height: 44px;`
-    let hideBtn = document.createElement('button')
+    let hideBtn = $el('button.closeBtn',{
+        innerHTML:closeIcon,
+        onclick:_=>div.style.display = 'none'
+    })
    let textB = document.createElement('p')
    btn.appendChild(textB)
    btn.appendChild(hideBtn)
    textB.style.fontSize = '11px'
    textB.innerHTML =  `<b>${$t('Groups Map')} (EasyUse)</b>`
-    hideBtn.style = `float: right;color: var(--input-text);border-radius:6px;font-size:9px;
-        background-color: var(--comfy-input-bg); border: 1px solid var(--border-color);cursor: pointer;padding: 5px;aspect-ratio: 1 / 1;`
-    hideBtn.addEventListener('click', () => {
-        div.style.display = 'none'
-    })
-    hideBtn.innerText = '❌'
    div.appendChild(btn)

    div.addEventListener('mousedown', function (e) {
@@ -301,10 +301,12 @@ function download_model(url,local_dir){
    })

 }
-class GuideDialog extends ComfyDialog {
-    show(note, need_models){
+class GuideDialog {
+
+    constructor(note, need_models){
+        this.dialogDiv = null
+        this.modelsDiv = null

-        let modelsDiv = null
        if(need_models?.length>0){
            let tbody = []

@@ -313,18 +315,17 @@ class GuideDialog extends ComfyDialog {
                    $el('td',{innerHTML:need_models[i].title || need_models[i].name || ''}),
                    $el('td',[
                        need_models[i]['download_url'] ? $el('a',{onclick:_=>download_model(need_models[i]['download_url'],need_models[i]['local_dir']), target:"_blank", textContent:$t('Download Model')}) : '',
-                         need_models[i]['source_url'] ? $el('a',{href:need_models[i]['source_url'], target:"_blank", textContent:$t('Source Url')}) : '',
+                        need_models[i]['source_url'] ? $el('a',{href:need_models[i]['source_url'], target:"_blank", textContent:$t('Source Url')}) : '',
+                        need_models[i]['desciption'] ? $el('span',{textContent:need_models[i]['desciption']}) : '',
                    ]),
-                    $el('td',{innerHTML:need_models[i].description || ''}),
                ]))
            }
-            modelsDiv = $el('div.easyuse-guide-dialog-models.markdown-body',[
+            this.modelsDiv = $el('div.easyuse-guide-dialog-models.markdown-body',[
                $el('h3',{textContent:$t('Models Required')}),
                $el('table',{cellpadding:0,cellspacing:0},[
                    $el('thead',[
                        $el('tr',[
                            $el('th',{innerHTML:$t('ModelName')}),
-                            $el('th',{innerHTML:$t('Details')}),
                            $el('th',{innerHTML:$t('Description')}),
                        ])
                    ]),
@@ -333,36 +334,61 @@ class GuideDialog extends ComfyDialog {
            ])
        }

-        super.show(
-           $el('div.easyuse-guide-dialog',[
-               $el('div.easyuse-guide-dialog-header',[
-                     $el('div.easyuse-guide-dialog-title',{
+        this.dialogDiv = $el('div.easyuse-guide-dialog.hidden',[
+           $el('div.easyuse-guide-dialog-header',[
+                 $el('div.easyuse-guide-dialog-top',[
+                    $el('div.easyuse-guide-dialog-title',{
                        innerHTML:$t('Workflow Guide')
-                     }),
-                     $el('div.easyuse-guide-dialog-remark',{
-                        innerHTML:`${$t('Workflow created by')} <a href="https://github.com/yolain/" target="_blank">Yolain</a> , ${$t('Watch more video content')} <a href="https://space.bilibili.com/1840885116" target="_blank">B站乱乱呀</a>`
-                     })
-               ]),
-               $el('div.easyuse-guide-dialog-content.markdown-body',[
-                   $el('div.easyuse-guide-dialog-note',{
-                       innerHTML:note
-                   }),
-                   modelsDiv
-               ])
+                    }),
+                    $el('button.closeBtn',{innerHTML:closeIcon,onclick:_=>this.close()})
+                 ]),
+
+                 $el('div.easyuse-guide-dialog-remark',{
+                    innerHTML:`${$t('Workflow created by')} <a href="https://github.com/yolain/" target="_blank">Yolain</a> , ${$t('Watch more video content')} <a href="https://space.bilibili.com/1840885116" target="_blank">B站乱乱呀</a>`
+                 })
+           ]),
+           $el('div.easyuse-guide-dialog-content.markdown-body',[
+               $el('div.easyuse-guide-dialog-note',{
+                   innerHTML:note
+               }),
+               ...this.modelsDiv ? [this.modelsDiv] : []
           ])
-        )
+        ])
+
+        if(disableRenderInfo){
+            this.dialogDiv.classList.add('disable-render-info')
+        }
+        document.body.appendChild(this.dialogDiv)
+    }
+    show(){
+        if(this.dialogDiv) this.dialogDiv.classList.remove('hidden')
    }

    close(){
-        guideDialog = null
-        super.close()
+        if(this.dialogDiv){
+            this.dialogDiv.classList.add('hidden')
+        }
+    }
+    toggle(){
+        if(this.dialogDiv){
+            if(this.dialogDiv.classList.contains('hidden')){
+                this.show()
+            }else{
+                this.close()
+            }
+        }
+    }
+
+    remove(){
+        if(this.dialogDiv) document.body.removeChild(this.dialogDiv)
    }
 }

+const getEnableToolBar = _ => app.ui.settings.getSettingValue(toolBarId, true)

 const toolBarId = "Comfy.EasyUse.toolBar"

-let enableToolBar = true
+let enableToolBar = getEnableToolBar()
 let disableRenderInfo = localStorage['Comfy.Settings.Comfy.EasyUse.disableRenderInfo'] ? true : false
 export function addToolBar(app) {
 	app.ui.settings.addSetting({
@@ -378,7 +404,6 @@ export function addToolBar(app) {
 		},
 	});
 }
-const getEnableToolBar = _ => app.ui.settings.getSettingValue(toolBarId, true)

 let note = null
 let toolbar = null
@@ -469,16 +494,15 @@ app.registerExtension({
            // }
            if(data?.extra?.note){
                 if(guideDialog) {
-                     guideDialog.close()
+                     guideDialog.remove()
                     guideDialog = null
                 }
                 if(note && toolbar) toolbar.removeChild(note)
+                 const need_models = data.extra?.need_models || null
+                 guideDialog = new GuideDialog(data.extra.note, need_models)
                 note = $el('div.easyuse-toolbar-item',{
                    onclick:async()=>{
-                        if(guideDialog) return
-                        guideDialog = new GuideDialog()
-                        const need_models = data.extra?.need_models || null
-                        guideDialog.show(data.extra.note, need_models)
+                        guideDialog.toggle()
                    }
                },[
                    $el('div.easyuse-toolbar-icon.question',{innerHTML:quesitonIcon}),
@@ -111,16 +111,31 @@ function widgetLogic(node, widget) {
 		updateNodeHeight(node)
 	}
 	if (widget.name === 'mode') {
-		let number_to_show = findWidgetByName(node, 'num_loras').value + 1
-		for (let i = 0; i < number_to_show; i++) {
-			if (widget.value === "simple") {
-				toggleWidget(node, findWidgetByName(node, 'lora_'+i+'_strength'), true)
-				toggleWidget(node, findWidgetByName(node, 'lora_'+i+'_model_strength'))
-				toggleWidget(node, findWidgetByName(node, 'lora_'+i+'_clip_strength'))
-			} else {
-				toggleWidget(node, findWidgetByName(node, 'lora_'+i+'_strength'))
-				toggleWidget(node, findWidgetByName(node, 'lora_'+i+'_model_strength'), true)
-				toggleWidget(node, findWidgetByName(node, 'lora_'+i+'_clip_strength'), true)}
+		switch (node.comfyClass) {
+			case 'easy loraStack':
+				let number_to_show = findWidgetByName(node, 'num_loras').value + 1
+				for (let i = 0; i < number_to_show; i++) {
+					if (widget.value === "simple") {
+						toggleWidget(node, findWidgetByName(node, 'lora_'+i+'_strength'), true)
+						toggleWidget(node, findWidgetByName(node, 'lora_'+i+'_model_strength'))
+						toggleWidget(node, findWidgetByName(node, 'lora_'+i+'_clip_strength'))
+					} else {
+						toggleWidget(node, findWidgetByName(node, 'lora_'+i+'_strength'))
+						toggleWidget(node, findWidgetByName(node, 'lora_'+i+'_model_strength'), true)
+						toggleWidget(node, findWidgetByName(node, 'lora_'+i+'_clip_strength'), true)}
+				}
+				break
+			case 'easy icLightApply':
+				if (widget.value === "Foreground") {
+					toggleWidget(node, findWidgetByName(node, 'lighting'), true)
+					toggleWidget(node, findWidgetByName(node, 'remove_bg'), true)
+					toggleWidget(node, findWidgetByName(node, 'source'))
+				} else {
+					toggleWidget(node, findWidgetByName(node, 'lighting'))
+					toggleWidget(node, findWidgetByName(node, 'source'), true)
+					toggleWidget(node, findWidgetByName(node, 'remove_bg'))
+				}
+				break
 		}
 		updateNodeHeight(node)
 	}
@@ -590,6 +605,7 @@ app.registerExtension({
 			case "easy rangeFloat":
 			case 'easy latentCompositeMaskedWithCond':
 			case 'easy pipeEdit':
+			case 'easy icLightApply':
 			case 'easy ipadapterApply':
 			case 'easy ipadapterApplyADV':
 			case 'easy ipadapterApplyEncoder':
@@ -12,6 +12,7 @@ const customPipeLineLink = "#7737AA"
 const customPipeLineSDXLLink = "#7737AA"
 const customIntLink = "#29699C"
 const customXYPlotLink = "#74DA5D"
+const customLoraStackLink = "#94dccd"
 const customXYLink = "#38291f"

 var customLinkColors = JSON.parse(localStorage.getItem('Comfy.Settings.ttN.customLinkColors')) || {};
@@ -20,6 +21,8 @@ if (!customLinkColors["PIPE_LINE_SDXL"] || !LGraphCanvas.link_type_colors["PIPE_
 if (!customLinkColors["INT"] || !LGraphCanvas.link_type_colors["INT"]) {customLinkColors["INT"] = customIntLink;}
 if (!customLinkColors["XYPLOT"] || !LGraphCanvas.link_type_colors["XYPLOT"]) {customLinkColors["XYPLOT"] = customXYPlotLink;}
 if (!customLinkColors["X_Y"] || !LGraphCanvas.link_type_colors["X_Y"]) {customLinkColors["X_Y"] = customXYLink;}
+if (!customLinkColors["LORA_STACK"] || !LGraphCanvas.link_type_colors["LORA_STACK"]) {customLinkColors["LORA_STACK"] = customLoraStackLink;}
+if (!customLinkColors["CONTROL_NET_STACK"] || !LGraphCanvas.link_type_colors["CONTROL_NET_STACK"]) {customLinkColors["CONTROL_NET_STACK"] = customLoraStackLink;}

 localStorage.setItem('Comfy.Settings.easyUse.customLinkColors', JSON.stringify(customLinkColors));

@@ -63,10 +63,10 @@ app.registerExtension({
                            selector.element.children[0].innerHTML = ''
                            if(method_values == 'selfie_multiclass_256x256'){
                                toggleWidget(this, findWidgetByName(this, 'confidence'), true)
-                                this.setSize([300, 200]);
+                                this.setSize([300, 260]);
                            }else{
                                toggleWidget(this, findWidgetByName(this, 'confidence'))
-                                this.setSize([300, 400]);
+                                this.setSize([300, 500]);
                            }
                            let list = getTagList(tags[method_values]);
                            selector.element.children[0].append(...list)
@@ -122,10 +122,10 @@ app.registerExtension({
                    }
                    if(method_values == 'selfie_multiclass_256x256'){
                        toggleWidget(this, findWidgetByName(this, 'confidence'), true)
-                        this.setSize([300, 200]);
+                        this.setSize([300, 260]);
                    }else{
                        toggleWidget(this, findWidgetByName(this, 'confidence'))
-                        this.setSize([300, 420]);
+                        this.setSize([300, 500]);
                    }
                },1)

@@ -92,10 +92,11 @@ async function displayImage(imgName, styleName) {
            img.src = empty_img
        }
    }
-    var x = e.pageX-pxy.x-100;
-    var y = e.pageY-pxy.y+25;
-    img.style.left =  x+"px";
-    img.style.top =  y+"px";
+    var scale = app?.canvas?.ds?.scale || 1;
+    var x = (e.pageX-pxy.x-100)/scale;
+    var y = (e.pageY-pxy.y+25)/scale;
+    img.style.left = x+"px";
+    img.style.top = y+"px";
    img.style.display = "block";
    img.style.borderRadius = "10px";
    img.style.borderColor = "var(--fg-color)"
@@ -47,7 +47,7 @@ class chooserImageDialog extends ComfyDialog {
    createButtons() {
        const btns = super.createButtons();
        btns[0].onclick = _ => {
-            cancelButtonPressed()
+            if (FlowState.running()) { send_cancel();}
            super.close()
        }
        btns.unshift($el('button', {
@@ -72,6 +72,10 @@ class chooserImageDialog extends ComfyDialog {
 function progressButtonPressed() {
    const node = app.graph._nodes_by_id[this.node_id];
    if (node) {
+        const selected = [...node.selected]
+        if(selected?.length>0){
+            node.setProperty('values',selected)
+        }
        if (FlowState.paused()) {
            send_message(node.id, [...node.selected, -1, ...node.anti_selected]);
        }
@@ -79,10 +83,26 @@ function progressButtonPressed() {
            skip_next_restart_message();
            restart_from_here(node.id).then(() => { send_message(node.id, [...node.selected, -1, ...node.anti_selected]); });
        }
+        const maxlength = node.imgs.length;
+        if (FlowState.paused_here(node.id) && selected>0) {
+            node.send_button_widget.name = (selected>1) ? "Progress selected (" + selected + '/' + maxlength  +")" : "Progress selected image";
+        } else if (FlowState.idle() && selected>0) {
+            node.send_button_widget.name = (selected>1) ? "Progress selected (" + selected + '/' + maxlength  +")" : "Progress selected image as restart";
+        }
+        else {
+            node.send_button_widget.name = "";
+        }
    }
 }

-function cancelButtonPressed() { if (FlowState.running()) { send_cancel(); } }
+function cancelButtonPressed() {
+    if (FlowState.running()) { send_cancel();}
+    const node = app.graph._nodes_by_id[this.node_id];
+    if (node) {
+        node.send_button_widget.name = "";
+        node.cancel_button_widget.name = "";
+    }
+}

 app.registerExtension({
    name:'comfy.easyuse.imageChooser',
@@ -126,9 +146,12 @@ app.registerExtension({
    },

    async nodeCreated(node, app) {
+
        if(node.comfyClass == 'easy imageChooser'){
            node.send_button_widget = node.addWidget("button", "", "", progressButtonPressed, {serialize: false});
            node.cancel_button_widget = node.addWidget("button", "", "", cancelButtonPressed, {serialize: false});
+            node.setProperty('values',[])
+
            /* Capture clicks */
            const org_onMouseDown = node.onMouseDown;

@@ -178,7 +201,7 @@ app.registerExtension({
                if (this.send_button_widget) {
                    this.send_button_widget.node_id = this.id;
                    const selection = ( this.selected ? this.selected.size : 0 ) + ( this.anti_selected ? this.anti_selected.size : 0 )
-                    const maxlength = this.imgs.length;
+                    const maxlength = this.imgs?.length || 0;
                    if (FlowState.paused_here(this.id) && selection>0) {
                        this.send_button_widget.name = (selection>1) ? "Progress selected (" + selection + '/' + maxlength  +")" : "Progress selected image";
                    } else if (FlowState.idle() && selection>0) {
Author	SHA1	Message	Date
yolain	7cbc2a2a2b	modify the pyproject.toml file	2024-05-23 13:33:03 +08:00
yolain	63811206c9	Merge pull request #182 from haohaocreates/publish Add Github Action for Publishing to Comfy Registry	2024-05-23 13:30:16 +08:00
yolain	48a4b5bfc6	Merge pull request #183 from haohaocreates/pyproject Add pyproject.toml for Custom Node Registry	2024-05-23 13:30:00 +08:00
yolain	7980f4eef4	fix:compatible with new brushnet versions #181	2024-05-23 12:33:45 +08:00
haohaocreates	b19c8b07ea	chore(pyproject): Add pyproject.toml for Custom Node Registry	2024-05-22 18:08:19 -04:00
haohaocreates	3f506d265c	chore(publish): Add Github Action for Publishing to Comfy Registry	2024-05-22 18:08:16 -04:00
yolain	d8af918e7d	fix:the modal for selecting an image can not be closed #180	2024-05-21 15:54:02 +08:00
yolain	33566e8474	fix:run error when input optional image in easy fullkSampler #178	2024-05-21 15:41:39 +08:00
yolain	ea0350c2dc	fix:preview image offset in easy stylesSelector when zooming the page #176	2024-05-18 15:35:28 +08:00
yolain	24526623fb	fix:iclight cache bug #173	2024-05-17 17:00:39 +08:00
yolain	7715ebfd06	Merge pull request #170 from chenpx976/main feat: optimize GPU memory management and model unloading	2024-05-16 12:52:09 +08:00
color	2ba814c131	feat: optimize GPU memory management and model unloading	2024-05-16 12:35:40 +08:00
yolain	d5ad332666	fix:icLightModel add to cache	2024-05-16 11:57:19 +08:00
yolain	fee7e7bf73	fix:some models were not successfully written to easyCache,resulting in slow secondary diffusion	2024-05-16 01:01:03 +08:00
yolain	01f17ff02b	remove:unnecessary import modules	2024-05-15 17:07:14 +08:00
yolain	da7120219a	fix:diffusers version>0.26.0 use different module #70	2024-05-15 10:50:19 +08:00
yolain	d616d18069	fix:compatible with cg-image-picker #169	2024-05-15 10:04:00 +08:00
yolain	dbf76f288c	Update README.md	2024-05-14 16:49:16 +08:00
yolain	05124006ba	fix:set_clip_options no longer compare version #165	2024-05-13 14:23:55 +08:00
yolain	4586af311c	fix:fooocus+dd wrong	2024-05-13 14:19:33 +08:00
yolain	1cea58c7cf	add:remove_bg in easy icLightApply	2024-05-11 02:13:55 +08:00
yolain	a84f7c4a58	fix:brushnet error in easy presamplingInpainting	2024-05-11 00:42:14 +08:00
yolain	1d9bf86560	add:easy icLightApply	2024-05-10 17:26:04 +08:00
yolain	8d352b85bc	add:easy imageSplitGrid	2024-05-09 16:08:03 +08:00
yolain	403562575c	add:easy imageCropFromMask & imageUncropFromBBOX	2024-05-06 12:52:37 +08:00
yolain	dbe2cd6569	fix:compatibility with comfyui-brushnet new commit	2024-05-05 11:30:41 +08:00
yolain	b3b0a961c5	fix:raise excenption when comfyui-brushnet is not installed	2024-05-04 19:26:18 +08:00
yolain	9dfb8b9c15	fix:layerdiffuse everything config #158	2024-05-04 19:19:01 +08:00
yolain	b4ea58946c	fix:supported brushnet for ays	2024-05-04 01:01:35 +08:00
yolain	513fc4b67e	support for brushnet model loading	2024-05-03 18:24:47 +08:00