• 版本 [version]
• 下载 196
• 文件大小 17588
• 文件计数 1
• 创建日期 2026年8月6日
• 最后更新 2026年9月21日

Wan2.2_Video_Upscale

Wan2.2_Video_Upscale 是一套面向视频生成场景的 ComfyUI 工作流，全图共 17 个节点、17 条连线。链路以 CLIPTextEncode ×2、VHS_VideoCombine ×1、VAEDecode ×1 为主干，输出视频，适合需要稳定批量出图的同学。主模型 4x-UltraSharp.pth、Wan2.2T2VWan2.2-T2V-A14B-LowNoise-Q6_K.gguf、Wanumt5_xxl_fp8_e4m3fn_scaled.safetensors。下载后直接导入即可使用，建议显存 16GB 起（视频生成建议 24GB）。

• 主模型：4x-UltraSharp.pth、Wan2.2T2VWan2.2-T2V-A14B-LowNoise-Q6_K.gguf、Wanumt5_xxl_fp8_e4m3fn_scaled.safetensors
• 核心节点：CLIPTextEncode ×2、VHS_VideoCombine ×1、VAEDecode ×1
• 输出形式：输出视频
• 规模：17 节点 / 17 连线
• 建议显存：16GB 起（视频生成建议 24GB）
• 适用场景：图生视频、首尾帧、数字人与长视频一致性控制工作流。

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）。

资源头图

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": "115b5f64-0734-4438-afcf-e7122530c4ec",
    "revision": 0,
    "last_node_id": 26,
    "last_link_id": 28,
    "nodes": [
        {
            "id": 12,
            "type": "CLIPTextEncode",
            "pos": [
                -235.20426777028183,
                823.7005178718115
            ],
            "size": [
                400,
                200
            ],
            "flags": [],
            "order": 8,
            "mode": 0,
            "inputs": [
                {
                    "label": "clip",
                    "localized_name": "clip",
                    "name": "clip",
                    "type": "CLIP",
                    "link": 14
                },
                {
                    "label": "text",
                    "localized_name": "text",
                    "name": "text",
                    "type": "STRING",
                    "widget": {
                        "name": "text"
                    },
                    "link": null
                }
            ],
            "outputs": [
                {
                    "label": "CONDITIONING",
                    "localized_name": "CONDITIONING",
                    "name": "CONDITIONING",
                    "type": "CONDITIONING",
                    "links": [
                        12
                    ]
                }
            ],
            "properties": {
                "cnr_id": "comfy-core",
                "ver": "0.3.49",
                "Node name for S&R": "CLIPTextEncode"
            },
            "widgets_values": [
                "Cinematic, Japanese clean aesthetic style. On a sun-drenched afternoon, aboard a train traveling along the coastline, an Asian woman with shoulder-length black hair, wearing a simple white dress, sits by the window. Bright light floods the carriage, and the sparkling sea is visible outside. Initially gazing intently at the fleeting scenery, she then turns her head, as if sensing something, and offers a gentle and hopeful smile towards the camera. The shot begins as a medium shot and slowly and steadily pushes in for a close-up on her face as her smile becomes clearer. Soft natural light bathes her face in a warm glow, and a gentle sea breeze seems to stir the strands of hair near her ears."
            ]
        },
        {
            "id": 24,
            "type": "VHS_VideoCombine",
            "pos": [
                1289.8241105912973,
                492.73580539106507
            ],
            "size": [
                226.5576171875,
                334
            ],
            "flags": [],
            "order": 16,
            "mode": 0,
            "inputs": [
                {
                    "label": "images",
                    "localized_name": "images",
                    "name": "images",
                    "type": "IMAGE",
                    "link": 28
                },
                {
                    "label": "audio",
                    "localized_name": "audio",
                    "name": "audio",
                    "shape": 7,
                    "type": "AUDIO",
                    "link": null
                },
                {
                    "label": "meta_batch",
                    "localized_name": "meta_batch",
                    "name": "meta_batch",
                    "shape": 7,
                    "type": "VHS_BatchManager",
                    "link": null
                },
                {
                    "label": "vae",
                    "localized_name": "vae",
                    "name": "vae",
                    "shape": 7,
                    "type": "VAE",
                    "link": null
                },
                {
                    "label": "frame_rate",
             
… (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。