- 版本 [version]
- 下载 200
- 文件大小 18333
- 文件计数 1
- 创建日期 2026年8月7日
- 最后更新 2026年9月20日
一键转绘 _ 灵魂画手(抖音热门)
一键转绘 _ 灵魂画手(抖音热门):基于 FLUX、ControlNet 搭建的线稿提取方案,节点规模 16 个、连线 18 条。工作流主路径为 LoraLoaderModelOnly ×1、ConditioningZeroOut ×1、VAEDecode ×1,配合 输出图像,默认输出 1024×1368;节点已按执行顺序排布,便于二次修改。主模型 FLUX.1-dev-ControlNet-Union-Pro-InstantX.safetensors。下载后直接导入即可使用,建议显存 12GB 起(建议 16GB 以上)。
- 主模型:FLUX.1-dev-ControlNet-Union-Pro-InstantX.safetensors
- 核心节点:LoraLoaderModelOnly ×1、ConditioningZeroOut ×1、VAEDecode ×1
- 输出形式:输出图像 · 1024×1368
- 规模:16 节点 / 18 连线
- 建议显存:12GB 起(建议 16GB 以上)
- 适用场景:线稿提取与上色,涂鸦/草图直出效果图,设计与建筑场景常用。
资源头图

工作流节点图
图较大可横向滚动;点击图片查看高清矢量原图(浏览器中可自由缩放)。
使用说明
下载工作流文件后在 ComfyUI 中直接拖入画布(或菜单 Load 导入)即可运行;缺失的自定义节点用 ComfyUI Manager 一键安装,模型按文件名放入对应的 models 目录。
源码预览
{
"id": "1ec65101-e974-4822-8935-16b882e4acf1",
"revision": 0,
"last_node_id": 65,
"last_link_id": 92,
"nodes": [
{
"id": 33,
"type": "LoraLoaderModelOnly",
"pos": [
-1205,
-46
],
"size": [
315,
82
],
"flags": [],
"order": 7,
"mode": 0,
"inputs": [
{
"label": "模型",
"name": "model",
"type": "MODEL",
"link": 56
}
],
"outputs": [
{
"label": "模型",
"name": "MODEL",
"type": "MODEL",
"slot_index": 0,
"links": [
57
]
}
],
"properties": {
"Node name for S&R": "LoraLoaderModelOnly"
},
"widgets_values": [
"抽象 | 灵魂画手(抖音热门)_v3.0",
0.8
],
"_widget_cache_map": {
"抽象 | 灵魂画手(抖音热门)_v3.0": {
"name": "抽象 | 灵魂画手(抖音热门)_v3.0",
"baseType": 19,
"value": "162d04fe6d8a483c85ffa1e7c0f7f71a"
}
}
},
{
"id": 31,
"type": "ConditioningZeroOut",
"pos": [
-1215.70654296875,
97.83700561523438
],
"size": [
317.4000244140625,
26
],
"flags": [],
"order": 10,
"mode": 0,
"inputs": [
{
"label": "条件",
"name": "conditioning",
"type": "CONDITIONING",
"link": 52
}
],
"outputs": [
{
"label": "条件",
"name": "CONDITIONING",
"type": "CONDITIONING",
"slot_index": 0,
"links": []
}
],
"properties": {
"Node name for S&R": "ConditioningZeroOut"
},
"widgets_values": []
},
{
"id": 8,
"type": "VAEDecode",
"pos": [
-1213.6510009765625,
171.009033203125
],
"size": [
210,
46
],
"flags": [],
"order": 13,
"mode": 0,
"inputs": [
{
"label": "Latent",
"name": "samples",
"type": "LATENT",
"link": 54
},
{
"label": "VAE",
"name": "vae",
"type": "VAE",
"link": 12
}
],
"outputs": [
{
"label": "图像",
"name": "IMAGE",
"type": "IMAGE",
"slot_index": 0,
"links": [
9,
89
]
}
],
"properties": {
"Node name for S&R": "VAEDecode"
},
"widgets_values": []
},
{
"id": 60,
"type": "ControlNetApplyAdvanced",
"pos": [
-828.8193359375,
120.74786376953125
],
"s
… (完整内容请下载文件查看)
代码较长,以上仅前 3800 字符 —— 点击下方按钮下载完整 JSON。



