百科.dev
全部条目AI 编程趋势榜开源项目技术资讯提交条目
登录
< 返回工具列表
C

ComfyUI-WanAnimatePlus

> 编程语言
开源

支持无缝视频连接和多参考图像,适用于 Wan Animate 和 Scail-2

390 stars0 点赞0 次浏览
访问官网GitHub

工具介绍

支持无缝视频连接和多参考图像,适用于 Wan Animate 和 Scail-2

ComfyUI-WanAnimatePlus

English | 中文

Multi-reference image injection and seamless video connection for ComfyUI's WanAnimate pipeline.

Overview

ComfyUI-WanAnimatePlus adds five major feature groups to the WanVideo workflow:

prefix_frames & transition_video

  • prefix_frames: allows passing 1–5 additional reference images for multi-reference guided generation
  • transition_video: allows passing the last 21 frames of the previous video segment for seamless video connection

When used together, canvas layout and frame offsets are automatically coordinated without conflicts.

Bernini

Supports Bernini models. Allows passing source video, reference video, or reference images as generation conditions. Supports v2v, rv2v, r2v, and t2v tasks.

Use cases:

  • Multi-shot video sequence generation
  • Video continuation / extension
  • Motion transfer with multi-reference control
  • Video editing with source video + reference images
  • Reference-to-video generation

SCAIL-2 Embeds

Adds a wrapper-native WanAnimatePlus SCAIL_2 Embeds node for SCAIL-2 models. It prepares reference image, driving pose, colored pose mask, reference mask, optional prefix/transition hard-freeze latents, and prefix-aligned colored masks for the WanAnimatePlus sampler.

Official-Compatible SCAIL-2 Flow Nodes

Adds WanAnimatePlus SCAIL_2 Flow Embeds, WanAnimatePlus SCAIL_2 Flow Sampler, and WanAnimatePlus VAE Decode. These nodes use official ComfyUI MODEL, VAE, CONDITIONING, LATENT, and IMAGE interfaces so they can fit into official-style workflows while preserving WanAnimatePlus SCAIL-2 prefix, transition, bg, mask, loop colormatch, and two-phase sampling behavior.

Official-Compatible Animate2 Embeds

Adds WanAnimatePlus Animate2 Embeds for official ComfyUI Animate2 models. It maps Plus-style ref_image, prefix_frames, and bg_image onto official Animate2 CONDITIONING (concat_latent_image, concat_mask, pose_video_latent) and LATENT. Sample with the same WanAnimatePlus SCAIL_2 Flow Sampler used by SCAIL-2 Flow; there is no separate Animate2 sampler.

Demo

RunningHub Online Workflow Examples

RunningHub is an online ComfyUI compute platform. If your local hardware cannot run AI models, you can run the workflows below for free on RunningHub; register through the links to receive 1000 credits.

Scail2 Ultra-long Video Two-stage Sampling Non-degradation Workflow Ultra-powerful all-in-one WanAnimate Plus version workflow
https://www.runninghub.ai/post/2073345711527444480/?inviteCode=rh-v1565 https://www.runninghub.ai/post/2073345712362110976/?inviteCode=rh-v1565

prefix_frames & transition_video usage

prefix_frames demo

transition_video demo

Features

prefix_frames (Multi-Reference Injection)

Allows 1–5 additional reference images. Internally expands the canvas pixel space and encodes reference images across the front frames, with automatic frame offset coordination for control signals (pose / face).

  • Supports 1–5 reference images (truncated if exceeding 5)
  • Auto-resizes reference images to target resolution
  • Automatically aligns frame offsets for pose / face / bg / mask signals

transition_video (Seamless Video Connection)

Allows passing the last 21 frames of the previous video segment. Writes these pixel frames directly into the front of the generation canvas, with sampled+reversed padding for control signal offsets.

  • Automatically coordinates with prefix when both are used

Bernini

Generates condition latents from source video, reference video, and/or reference images via VAE encoding. Supports v2v, rv2v, r2v, and t2v — task is auto-detected from connected inputs.

  • Reference images kept at native aspect ratio
  • Compatible with context windows

SCAIL-2

Provides SCAIL-2 ref / pose / mask conditioning through WanAnimatePlus SCAIL_2 Embeds.

  • Encodes ref_image, pose_images, pose_image_mask, prefix_frames, prefix_mask, bg_image, and reference_image_mask
  • Aligns SCAIL-2 inputs to 32-pixel multiples before VAE encoding
  • Supports animation and replacement modes
  • Supports single-frame prefix reference encoding and optional transition_video hard-freeze conditioning
  • By default, prefix_frames are encoded as full-resolution reference latents and do not expand the output canvas; disable single_frame_prefix_encoding to use the legacy 37 front pixel-frame prefix layout
  • In single-frame prefix mode, prefix_mask follows the same reference-mask path as reference_image_mask
  • Supports context-window sampling; non-first windows can see prepended prefix/transition context without fusing those prepended predictions

Animate2

Provides official Animate2 conditioning through WanAnimatePlus Animate2 Embeds, sampled by WanAnimatePlus SCAIL_2 Flow Sampler.

  • Maps ref_image, prefix_frames, and bg_image onto official concat_latent_image / concat_mask / pose_video_latent
  • prefix_frames are extra frozen identity latents (max 5) and are trimmed after sampling
  • bg_image fills the unknown canvas instead of mid-grey and remains generated
  • Internal looping uses a 5-frame concat/mask handoff when frame_window_size is smaller than num_frames
  • Does not wrap official WanAnimate2Cache / PoseBranchCache; use those nodes directly when caching is needed

Installation

Place this repository into ComfyUI's custom_nodes directory:

cd ComfyUI/custom_nodes
git clone https://github.com/wuwukaka/ComfyUI-WanAnimatePlus.git

Restart ComfyUI after installation.

Important: To use prefix_frames, transition_video, Bernini, or SCAIL_2 Embeds, you must replace the full workflow chain with WanAnimatePlus nodes. Mixing WanAnimatePlus nodes with original WanVideoWrapper nodes in the same workflow will result in degraded output.

Quick Start

  1. Start ComfyUI and confirm the WanAnimatePlus nodes appear under the WanAnimatePlus category
  2. Replace the entire workflow chain with WanAnimatePlus counterparts: ModelLoader, VAELoader, ContextOptions, AnimateEmbeds, Sampler, Decode, and supporting nodes
  3. Do not mix original WanVideoWrapper nodes in the same workflow
  4. For SCAIL-2 and WanAnimate workflows, WanAnimatePlus Easy Sampler or WanAnimatePlus Easy SamplerSettings is recommended because they keep the full sampler functionality while exposing only the common controls
  5. For official ComfyUI MODEL/VAE/CONDITIONING/LATENT/IMAGE chains:
    • SCAIL-2: WanAnimatePlus SCAIL_2 Flow Embeds -> WanAnimatePlus SCAIL_2 Flow Sampler -> WanAnimatePlus VAE Decode
    • Animate2: WanAnimatePlus Animate2 Embeds -> WanAnimatePlus SCAIL_2 Flow Sampler -> WanAnimatePlus VAE Decode
  6. Connect prefix_frames and/or transition_video inputs as needed
  7. Example workflows are available in the example_workflows/ directory

Nodes

WanAnimatePlus exposes a complete workflow chain to avoid cross-package object mixing with the original WanVideoWrapper nodes.

Core nodes:

  • WanAnimatePlus ModelLoader
  • WanAnimatePlus VAELoader
  • WanAnimatePlus TextEncodeCached
  • WanAnimatePlus ClipVisionEncode
  • WanAnimatePlus ContextOptions
  • WanAnimatePlus AnimateEmbeds
  • WanAnimatePlus Sampler / WanAnimatePlus Samplerv2
  • WanAnimatePlus Easy Sampler / WanAnimatePlus Easy SamplerSettings
  • WanAnimatePlus Scheduler / WanAnimatePlus Schedulerv2
  • WanAnimatePlus Decode / WanAnimatePlus Encode
  • WanAnimatePlus LoraSelect / WanAnimatePlus LoraSelectMulti / WanAnimatePlus SetLoRAs
  • WanAnimatePlus BlockSwap / WanAnimatePlus SetBlockSwap
  • WanAnimatePlus TorchCompileSettings
  • WanAnimatePlus SamplerExtraArgs
  • WanAnimatePlus Uni3C ControlnetLoader / WanAnimatePlus Uni3C Embeds
  • WanAnimatePlus Bernini
  • WanAnimatePlus SCAIL_2 Embeds
  • WanAnimatePlus SCAIL_2 Flow Embeds
  • WanAnimatePlus SCAIL_2 Flow Sampler
  • WanAnimatePlus Animate2 Embeds
  • WanAnimatePlus VAE Decode

WanAnimatePlus Easy Sampler / Easy SamplerSettings

These nodes are recommended for SCAIL-2 and WanAnimate workflows. They simplify the visible node panel to common controls such as steps, cfg, shift, seed, force_offload, and scheduler, while still using the full WanAnimatePlus sampler settings and feature path underneath.

WanAnimatePlus Easy Sampler runs sampling directly. WanAnimatePlus Easy SamplerSettings outputs SAMPLER_ARGS for use with WanAnimatePlus SamplerFromSettings.

WanAnimatePlus AnimateEmbeds

Core node, replaces the original WanVideoAnimateEmbeds.

New inputs:

Input Description
prefix_frames 1–5 additional reference images for multi-reference guided generation
transition_video Last 21 frames of the previous video segment for seamless video connection

Other inputs are identical to the original WanVideoAnimateEmbeds: vae, width, height, num_frames, ref_images, pose_images, face_images, bg_images, mask, start_ref_image, clip_embeds, etc.

WanAnimatePlus Bernini

Generates condition latents from source video, reference video, and/or reference images for Bernini models.

Inputs:

Input Description
vae VAE model for encoding
width / height / num_frames Output dimensions
source_video Source video to edit/restyle (v2v/rv2v). Resized to width/height
reference_video Moving content to composite (video insertion), native aspect
reference_images Reference image(s) as in-context tokens (r2v/rv2v). Native aspect
ref_max_size Max long-edge size for reference media (default 848)
force_offload Offload VAE after encoding to save VRAM
tiled_vae Use tiled VAE encoding for memory savings

The task (v2v, rv2v, r2v, t2v) is automatically inferred from which inputs are connected.

WanAnimatePlus SCAIL_2 Embeds

Creates SCAIL-2 conditioning for WanAnimatePlus sampling. Use this node with SCAIL-2 checkpoints that include the pose and mask streams.

Inputs:

Input Description
vae VAE model for encoding
width / height / num_frames Target dimensions; width and height are aligned to multiples of 32
ref_image Reference image for SCAIL-2 conditioning
bg_image Optional single background image for animation mode. In single-frame mode it is encoded as an extra background reference latent; in legacy canvas-prefix mode it is appended after user prefix_frames. Ignored in replacement mode
pose_images Driving pose video/images, encoded at half resolution
pose_image_mask Colored per-identity pose mask sequence
prefix_mask Optional colored mask images matching prefix_frames; in single-frame mode it follows the reference-mask path, and in legacy canvas-prefix mode it is expanded as 1+4+4... and written into prefix pixel mask frames
reference_image_mask Colored reference mask image
replacement_mode Enables SCAIL-2 replacement-mode RoPE and reference-mask compositing
preserve_main_ref_background Animation mode only; keeps the main reference image background when enabled, or uses reference_image_mask as a black-background alpha crop when disabled. Ignored in replacement mode
single_frame_prefix_encoding Encodes prefix_frames as individual full-resolution reference latents instead of expanding the canvas; enabled by default
prefix_frames Optional prefix images. In default single-frame mode these become reference-stream latents; with single-frame mode disable

Issues· 0 开放

查看全部 Issues在 GitHub 打开

暂无开放 Issues，或尚未同步最近议题。

> 标签

Python

暂无评论，来聊聊你的看法吧

> 工具信息

发布日期2026年8月1日
最后更新2026年9月17日
分类编程语言
定价开源

> 相关工具

T
TypeScript
JavaScript 的超集，为前端与全栈提供静态类型
P
Python
通用编程语言，广泛用于 Web、数据与 AI
G
Go
Google 推出的简洁高效系统语言