• 版本 [version]
• 下载 144
• 文件大小 22223
• 文件计数 1
• 创建日期 2025年9月18日
• 最后更新 2026年9月21日

ltxv_0.95_i2v

ltxv_0.95_i2v 是一套面向视频生成场景的 ComfyUI 工作流，全图共 20 个节点、28 条连线。链路以 CLIPTextEncode ×2、SamplerCustomAdvanced ×1、LTXVImgToVideo ×1 为主干，输出视频，适合需要稳定批量出图的同学。主模型 t5xxl_fp8_e4m3fn.safetensors、LTXVltx-video-2b-v0.9.5.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：t5xxl_fp8_e4m3fn.safetensors、LTXVltx-video-2b-v0.9.5.safetensors
• 核心节点：CLIPTextEncode ×2、SamplerCustomAdvanced ×1、LTXVImgToVideo ×1
• 输出形式：输出视频
• 规模：20 节点 / 28 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": "7c8a248c-374b-4a1c-8ce0-fc5215eb2183",
    "revision": 0,
    "last_node_id": 91,
    "last_link_id": 221,
    "nodes": [
        {
            "id": 81,
            "type": "SamplerCustomAdvanced",
            "pos": [
                1199.310302734375,
                17.57975959777832
            ],
            "size": [
                355.20001220703125,
                106
            ],
            "flags": [],
            "order": 17,
            "mode": 0,
            "inputs": [
                {
                    "label": "noise",
                    "localized_name": "噪波",
                    "name": "noise",
                    "type": "NOISE",
                    "link": 194
                },
                {
                    "label": "guider",
                    "localized_name": "引导器",
                    "name": "guider",
                    "type": "GUIDER",
                    "link": 200
                },
                {
                    "label": "sampler",
                    "localized_name": "采样器",
                    "name": "sampler",
                    "type": "SAMPLER",
                    "link": 196
                },
                {
                    "label": "sigmas",
                    "localized_name": "西格玛",
                    "name": "sigmas",
                    "type": "SIGMAS",
                    "link": 197
                },
                {
                    "label": "latent_image",
                    "localized_name": "Latent图像",
                    "name": "latent_image",
                    "type": "LATENT",
                    "link": 198
                }
            ],
            "outputs": [
                {
                    "label": "output",
                    "localized_name": "Latent",
                    "name": "output",
                    "type": "LATENT",
                    "slot_index": 0,
                    "links": [
                        216
                    ]
                },
                {
                    "label": "denoised_output",
                    "localized_name": "降噪Latent",
                    "name": "denoised_output",
                    "type": "LATENT",
                    "links": null
                }
            ],
            "properties": {
                "cnr_id": "comfy-core",
                "ver": "0.3.29",
                "Node name for S&R": "SamplerCustomAdvanced"
            },
            "widgets_values": []
        },
        {
            "id": 77,
            "type": "LTXVImgToVideo",
            "pos": [
                865.2247924804688,
                376.3932189941406
            ],
            "size": [
                315,
                230
            ],
            "flags": [],
            "order": 13,
            "mode": 0,
            "inputs": [
                {
                    "label": "positive",
                    "localized_name": "正面条件",
                    "name": "positive",
                    "type": "CONDITIONING",
                    "link": 187
                },
                {
                    "label": "negative",
                    "localized_name": "负面条件",
                    "name": "negative",
                    "type": "CONDITIONING",
                    "link": 188
                },
                {
                    "label": "vae",
                    "localized_name": "vae",
                    "name": "vae",
                    "type": "VAE",
                    "link": 213
                },
                {
                    "label": "image",
                    "localized_name": "图像",
                    "name": "image",
                    "type": "IMAGE",
                    "link": 204
                },
           
… (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。