📦 🎬 Video Edit — Pro Pack on RunComfy — 🎬 Video Edit — Pro Pack on 运行Comfy

🎬 Video Edit — Pro Pack on 运行Comfy

运行comfy.com · docs · Wan 2.7 Edit-Video · Kling Motion-Control Pro · Lucy Edit Restyle

Video edit, intent-路由d. This 技能 doesn't lock you to one 模型 — it picks the right video-edit 模型 in the 运行Comfy cata记录 based on what the user actually wants: general restyle, motion transfer from a reference 命令行工具p, or lightweight 身份-stable outfit / background swap.

User intent 模型 Why Restyle a talking-head video — preserve face / pose / lip movement Wan 2.7 Edit-Video Strong 身份 + motion preservation; supports up to 1080p Swap product background, keep camera motion Wan 2.7 Edit-Video Camera motion preserved; one-direction edit honored Replace packaging de签名 using a reference image Wan 2.7 Edit-Video + reference_image Reference-conditioned de签名 transfer 应用ly cinematic color grade / commercial polish Wan 2.7 Edit-Video Good at single-direction global look changes Transfer precise motion from a reference video to a tar获取 character Kling 2.6 Pro Motion Control De签名ed for motion m应用ing with 身份 hold Lip-同步 motion of a tar获取 character to source video's lip movement Kling 2.6 Pro Motion Control Built for tight temporal coherence Lightweight outfit / costume swap with 身份 preservation Lucy Edit Restyle Core strength is localized 身份-stable edits 身份-stable restyle ("astronaut in desert", "warm golden-hour lighting") Lucy Edit Restyle Specializes in temporal consistency for restyle Default if unspecified Wan 2.7 Edit-Video Most versatile, highest resolution

The 代理 reads this table, classifies the user's intent, and picks the matching subsection below.

Prerequisites 运行Comfy 命令行工具 — npm i -g @运行comfy/命令行工具 运行Comfy account — 运行comfy 记录in. CI / contAIners — 设置 运行COMFY_令牌=<令牌>. A source video URL — 格式化s and limits depend on the chosen 路由. 路由 1: Wan 2.7 Edit-Video — default for restyle / background / packaging

模型: wan-AI/wan-2-7/edit-video

模式 Field Type Required Default Notes prompt string yes — Lead with preservation. One edit direction per call. video string yes — MP4/MOV URL, 2–10s, ≤100MB. reference_image string no — URL — use for direct de签名 / 应用earance transfer only. resolution enum no (输入) 720p or 1080p. aspect_ratio enum no (输入) W:H. Defaults to 输入. duration int no 0 0 = match 输入; 2–10 = t运行cate from 启动. audio_设置ting enum no auto auto or origin (preserve source audio). 种子 int no — Reproducibility. Invoke

Background swap, 身份 preserved, audio kept:

运行comfy 运行 wan-AI/wan-2-7/edit-video \ --输入 '{ "prompt": "Preserve the speaker'\''s face, pose, and lip movement; change the background to a modern office with neutral lighting.", "video": "https://.../speaker.mp4", "audio_设置ting": "origin" }' \ --输出-dir

Packaging swap with reference image:

运行comfy 运行 wan-AI/wan-2-7/edit-video \ --输入 '{ "prompt": "MAIntAIn the original framing and hand movement; replace the packaging de签名 using the reference image.", "video": "https://.../hand-holding-package.mp4", "reference_image": "https://.../new-packaging.png", "audio_设置ting": "origin" }' \ --输出-dir

Prompting tips Preservation goals first: "Preserve [face / pose / motion / framing / lip movement]; [then 状态 the change]". One edit direction per call. Compound edits drift on motion. reference_image only when justified (packaging swap, costume swap with tar获取 visual). Don't pass refs for general restyle. audio_设置ting: "origin" for talking-head where you don't want sound追踪 re生成d. Source video constrAInts: 2–10s, ≤100MB. 路由 2: Kling 2.6 Pro Motion Control — when motion FROM a reference 命令行工具p is the point

模型: kling/kling-2-6/motion-control-pro

Use when the user wants to transfer the motion of a reference video onto a tar获取 character (driven by an image OR another video). This isn't restyle — it's motion m应用ing with 身份 hold.

模式 Field Type Required Notes prompt string yes Describe tar获取 motion / style. image string yes (image orientation) Reference for character / background consistency. video string yes Motion reference. 10–30s depending on orientation. keep_original_sound bool no Preserve audio from reference video. character_orientation enum yes image (max 10s 输出) or video (max 30s 输出). Invoke 运行comfy 运行 kling/kling-2-6/motion-control-pro \ --输入 '{ "prompt": "A young american woman dancing", "image": "https://.../tar获取-character.jpg", "video": "https://.../motion-reference-dance.mp4", "character_orientation": "image", "keep_original_sound": true }' \ --输出-dir

Prompting tips Subject must be > 5% of frame in the image reference for 清理 身份 hold. Spatial constrAInts help: "character on left side, background motion right". Simplify if 结果s drift between iterations — drop adjectives, keep core motion description. character_orientation: "image" caps outp