Extensions/XB_ToolBox
ComfyUI Extension

XB_ToolBox

A comprehensive ComfyUI extension suite encompassing both front-end interaction and low-level memory scheduling, designed to help beginners master workflows and simplify local deployment.

By wjluoxiao·Created 7 months ago·Updated about 17 hours ago· 462
WJLUOXIAO/XB_ToolBox
Nodes143
On cloudLocal install
CategoryXB_ToolBox/Utils, XB_ToolBox/Audio
Stars462
Updatedabout 17 hours ago

Nodes (143)

XB-BOX - 🔀 切换任意

The switch with no switch that just grabs the first input that isn't empty

XB_ToolBox/Utils
XB-BOX - 🎵 音频切片(基础)

Slice audio to exact frames for video models — no manual math

XB_ToolBox/Audio
XB-BOX - 🎵 音频切片V1(单人)

The audio slicer with a waveform you can actually click on

XB_ToolBox/Audio
XB-BOX - 🎵 音频切片V2(双人)

Stitch two voice clips into one track, gap and all

XB_ToolBox/Audio
XB-BOX - 🎵 音频切片V3(高级)

The audio slicer for when two voices actually talk over each other

XB_ToolBox/Audio
XB-BOX - 🤖 漫画文字渲染 (全自动带涂改液)

Erase AI gibberish in comic bubbles and typeset real dialogue

XB_ToolBox/Comic
XB-BOX - 📂 图片批量加载
XB_ToolBox/Batch_Process
XB-BOX - 📂 图片序列加载

Load one image at a time from a folder, stepper built in

XB_ToolBox/Batch_Process
XB-BOX - 🖼️ 批量图像

Merge loose images into one batch tensor, mute-proof

XB_ToolBox/Utils
XB-BOX - 🎨 Bernini 条件注入(VAE分块)

Feed source video and reference images into Bernini without OOMing

XB_ToolBox/Wan
XB-BOX - 🧠 Bernini 提示词路由(超级节点)

One dropdown replaces a 30-node Bernini prompt-switching subgraph

XB_ToolBox/Pipeline
XB-BOX - 🏷️ Canvas Label (文字标签)

A sticky note for your ComfyUI graph that can't break anything

XB_ToolBox/Display
XB-BOX - Checkpoint Block Swap

Push transformer blocks to RAM so big video models fit on your GPU

XB_ToolBox/VRAM_Hacks
XB-BOX - Chunk Visualization

See your spatial and temporal chunks before you blow your VRAM budget

XB_ToolBox
XB-BOX - CLIP Name Broadcaster

Turn a CLIP model name into a wire so other nodes can read it

XB_ToolBox/Wiring
XB-BOX - 📝 CLIP 文本编辑

Type the prompt once, get a positive and a kill-switch negative

XB_ToolBox/Text
XB-BOX - 📝 漫画提示词智能解析

Strip dialogue out of your comic prompt before the model sees it

XB_ToolBox/Comic
XB-BOX - 💬 漫画文字渲染 (精确坐标)

Place comic text at exact coordinates — narration, SFX, anything

XB_ToolBox/Comic
XB-BOX - ✂️ CosyVoice3 音频裁剪

Trim a reference clip by wall-clock time, not samples

🔊XB CosyVoice3/Audio
XB-BOX - 🌐 CosyVoice3 跨语言合成

Clone a voice once, then make it speak any of ten languages

🔊XB CosyVoice3/Synthesis
XB-BOX - 💬 CosyVoice3 多人对话

Two cloned voices, one script, one mixed conversation track

🔊XB CosyVoice3/Synthesis
XB-BOX - 🎭 CosyVoice3 指令控制

Tell the cloned voice how to sound, in plain words

🔊XB CosyVoice3/Synthesis
XB-BOX - 🔊 CosyVoice3 模型加载

The CosyVoice3 loader that actually gets the model into ComfyUI

🔊XB CosyVoice3/Loaders
XB-BOX - 💾 CosyVoice3 保存说话人

Turn a voice clip into a reusable .pt speaker file, once

🔊XB CosyVoice3/Utilities
XB-BOX - 🗣️ CosyVoice3 说话人克隆

Synthesize in a saved voice, no reference audio required

🔊XB CosyVoice3/Synthesis
XB-BOX - 🎭 CosyVoice3 说话人指令

Same cloned voice, but now tell it how to say the line

🔊XB CosyVoice3/Synthesis
XB-BOX - 🔄 CosyVoice3 语音转换

Respeak someone else's words in your own voice

🔊XB CosyVoice3/Synthesis
XB-BOX - 🎤 CosyVoice3 零样本克隆

Clone a voice from one audio clip, no training, no transcript required

🔊XB CosyVoice3/Synthesis
XB-BOX - 🤖 数字人参数调节(双人)

Two voices, two audio tracks, one synchronized frame count

XB_ToolBox/DigitalHuman
XB-BOX - 🤖 数字人参数调节(单人)

Frame math and audio slicing for a one-person talking head, in one node

XB_ToolBox/DigitalHuman
XB-BOX - 🚫 禁用分组

One switch strip that turns a whole block of your workflow off

XB_ToolBox/Utils
XB-BOX - 🚫 禁用节点

Mute, bypass and restore specific nodes without hunting for them

XB_ToolBox/Utils
XB-BOX - 🖼️ FLUX多图编码

Stuff up to nine reference images into a Flux conditioning, in one node

XB_ToolBox/FLUX
XB-BOX - 🌊 海螺H3视频参数

Hailuo H3 resolution and frame math, without the mental arithmetic

XB_ToolBox/Image_Params
XB-BOX - ✂️ 人物分割 (DirectML/ROCm)

Cut a person out of any image with a GPU-accelerated mask

XB_ToolBox/Segment
XB-BOX - 📥 人物分割模型加载

Load the ONNX segmentation model that the segmenter needs

XB_ToolBox/Segment
XB-BOX - Image Params Master

One hub for image size, batch, and strength — with dims that can't go invalid

XB_ToolBox/Image_Params
XB-BOX - Image Params Master Mini

One place to set the canvas — and it snaps to the grid for you

XB_ToolBox/Image_Params
XB-BOX - 🖼️ 生图提示词预设(图)
XB_ToolBox/Image_Params
XB-BOX - Image Prompt Preset

Prompt, Ratio and Empty Latent in XB-BOX

XB_ToolBox/Image_Params
XB-BOX - 🖼️ 生图提示词预设Pro

Prompt presets, a correct empty latent, and an optional LLM — in one node

XB_ToolBox/Image_Params
XB-BOX - 📐 批量缩放图像

Scale up to nine images at once with one shared setting

XB_ToolBox/Utils
XB-BOX - 📐 批量缩放图像(像素)
XB_ToolBox/Utils
XB-BOX - 🎨 K2 风格切换路由(超级节点)

One dropdown instead of ten toggles for your K2 style switch

XB_ToolBox/Pipeline
XB-BOX - 采样器(原版优化)

The stock KSampler, plus a before-sampling VRAM cleanup knob

XB_ToolBox/原版优化
XB-BOX - 高级采样器(原版优化)

KSamplerAdvanced, with a pre-run VRAM cleanup bolted on

XB_ToolBox/原版优化
XB-BOX - 📋 列表分发

Split one list of strings into N separate text wires

XB_ToolBox/Utils
XB-llama - 🔍 BBoxes取BBox

Pluck one box out of a list of LLM-detected boxes

XB-llama
XB-llama - 🧹 清理状态

Reset the LLM's memory so it stops 'remembering' old runs

XB-llama
XB-llama - 💬 指令推理

Run your local GGUF model as a full instruct + vision node

XB-llama
XB-llama - ✨ MiniMax-fl2va提示词预设

A complete MiniMax H3 prompt builder in one dropdown

XB-llama
XB-llama - ✨ MiniMax-ref2va提示词预设

The multi-reference MiniMax H3 prompt preset

XB-llama
XB-llama - 📦 模型加载器

The GGUF model loader that powers the whole XB-llama stack

XB-llama
XB-llama - 🚀 模型加载Pro

XB_ToolBox's GGUF loader doesn't load anything — and that's the feature

XB-llama
XB-llama - ⚙️ 推理参数

The sampling knobs for your local LLM, tuned for storyboards

XB-llama
XB-llama - ✨ 提示词增强预设

Model-specific system prompts, one dropdown at a time

XB-llama
XB-llama - ✨ 提示词增强反推

The prompt writer without the rest — local or API, text or image

XB-llama
XB-llama - ✨ 分镜增强预设

Turn a story idea into a frame-by-frame storyboard prompt

XB-llama
XB-llama - 💬 分镜推理

Story idea in, frame prompts out

XB-llama
XB-llama - 🎞️ 分镜词处理器

Draw cards from your storyboard, one shot at a time

XB-llama
XB-llama - 🎞️ 分镜词处理器Pro

Same storyboard gacha, plus it sniffs your characters for you

XB-llama
XB-llama - 🗑️ 卸载模型

Kick the LLM off your GPU when you're done with it

XB-llama
XB-llama - 📝 解包代码块

Strip the ``` fences off your LLM's JSON

XB-llama
XB-BOX - 🔄 LTX2.3 数字人无限时长对口型

Unlimited-length LTX 2.3 lipsync, stitched from segments

XB_ToolBox/Pipeline
XB-BOX - Master Parameter

One value box that snap-clamps itself to the mode you're in

XB_ToolBox/Params_Control
XB 📊 硬件监控面板 (Memory Visualization)

The VRAM radar that parks itself in the corner of your screen

XB_ToolBox/Dashboard
XB-BOX - 📦 模型加载大全V1 (GGUF)

V1's loader, but the model dropdown speaks GGUF

XB_ToolBox/Model_Loader_GGUF
XB-BOX - 📦 模型加载大全V2 (GGUF)

High/low dual-model loader, GGUF edition

XB_ToolBox/Model_Loader_GGUF
XB-BOX - 📦 模型加载大全V3 (GGUF)

LTX 2.3's dual-CLIP, dual-VAE stack, quantized to fit

XB_ToolBox/Model_Loader_GGUF
XB-BOX - 🎞️ MSR 多图合成帧序列

Four images in, one frame sequence out — MSR for short dramas

XB_ToolBox/Image_Params
XB-BOX - ⚡ 节点状态开关

Mute or bypass whole nodes from a boolean — without right-clicking

XB_ToolBox/Utils
XB-BOX - 🔊 播放声音

The beep that tells you the queue is finally done

XB_ToolBox/Utils
XB-BOX - 🖼️ Qwen2.1提示词预设

The text encoder you already loaded is a working LLM

XB_ToolBox/Image_Params
XB-BOX - 🔗 引用任意

One output, many optional upstreams — pick the live one by name

XB_ToolBox/Utils
XB-llama - 🎬 角色场景调度器

Turn one storyboard script into per-scene prompts plus matching characters

XB-llama
XB-BOX - Acceleration Mode Selector

Free attention speedup, with a preset dropdown for your exact GPU

XB_ToolBox/VRAM_Hacks
XB-BOX - Acceleration + BlockSwap Duo

The 'golden duo' that runs big models on small cards

XB_ToolBox/VRAM_Hacks
XB-BOX - 自定义采样器(原版优化)

A sampler that does exactly what it says, plus a VRAM shovel

XB_ToolBox/原版优化
XB-BOX - 自定义高级采样器(原版优化)

ComfyUI's most flexible sampler, gift-wrapped with a cleanup dropdown

XB_ToolBox/原版优化
XB-BOX - 🎬 保存视频

The two-node video ending, welded into one

XB_ToolBox/Video
XB-BOX - 🖌️ ShenBi Brush

A drawing pad that lives inside your ComfyUI graph

XB_ToolBox/Image
XB-BOX - Storyboard Slicer

One click from storyboard grid to individual scene frames

XB_ToolBox/Pipeline
XB-BOX - 📝 字符串合并

Join N strings without a tangle of text nodes

XB_ToolBox/Text
XB-BOX - 📝 Qwen图像编辑编码

Tell a picture what to change, no mask required

XB_ToolBox/Qwen
XB-BOX - 📝 Qwen图像编辑编码Plus

Multi-image Qwen editing, with the vision-language plumbing done for you

XB_ToolBox/Qwen
XB-BOX - 🔗 三连 LoRA 加载

A permanent three-slot LoRA rack for your graph

XB_ToolBox/原版优化
XB-BOX - 🔗 三连 LoRA 加载(仅模型)

For the models that don't want the text encoder touched

XB_ToolBox/原版优化
XB-BOX - UNet Block Swap

Trading speed for VRAM, one transformer block at a time

XB_ToolBox/VRAM_Hacks
XB-BOX - UNet Name Broadcaster

One dropdown, one wire, every model name you need

XB_ToolBox/Wiring
XB-BOX - VAE解码(原版优化)

The plainest node in the pack, and that's the point

XB_ToolBox/原版优化
XB-BOX - VAE分块解码(原版优化)

The decoder you actually want on a 16GB card

XB_ToolBox/原版优化
XB-BOX - VAE分块解码·图片(原版优化)

Decode huge latents without OOM — this is ComfyUI's tiled decoder with a cleanup switch

XB_ToolBox/原版优化
XB-BOX - VAE编码(原版优化)

Stock VAE encode, with a VRAM cleanup knob you can actually see

XB_ToolBox/原版优化
XB-BOX - VAE修补编码(原版优化)

Inpainting encode with a cleanup knob — same masked latent, less OOM

XB_ToolBox/原版优化
XB-BOX - VAE分块编码(原版优化)

Tiled VAE encode for big images and video — with the cleanup switch ComfyUI forgot

XB_ToolBox/原版优化
XB-BOX - VAE分块编码·图片(原版优化)
XB_ToolBox/原版优化
XB-BOX - 🎬 视频拼接输出

The VHS video saver that got ported into XB_ToolBox — with GIF/WebP shortcuts

XB_ToolBox/Video
XB-BOX - 🎬 视频加载器(修复预览BUG)

Load any video into frames with VHS compatibility — and the preview bug is actually fixed

XB_ToolBox/Video
XB-BOX - Video Merger

Stitch up to ten video segments into one long clip — on the CPU, so it won't OOM

XB_ToolBox/Pipeline
XB-BOX - Video Params Master

One hub for width, height, frames and FPS — with the 1+8N and golden-bucket rules built in

XB_ToolBox/Image_Params
XB-BOX - VRAM Calculator

Does your card actually fit that 22B model? This node does the math so you don't have to

XB_ToolBox/Chunk_Preview
XB_Wan22FunControlToVideo

Wan 2.2 reference + control-video conditioning, prebuilt so you don't hand-build concat latents

XB_ToolBox/Pipeline
XB_Wan22ImageToVideoLatent

A start image in, a Wan 2.2 latent out — the I2V setup that doesn't fight the VAE

XB_ToolBox/Pipeline
XB-BOX - 🎬 Animate 动作迁移总线

One node to configure a whole motion-transfer run — the Animate parameter bus

XB_ToolBox/Pipeline
XB-BOX - 🏃‍♀️ Animate 无限接力点

Chaining motion transfer past the 81-frame wall — the Animate relay node

XB_ToolBox/Pipeline
XB-BOX - 🆕 Animate 无限接力点 (New)

The better relay — one node that loops as many Animate segments as you need

XB_ToolBox/Pipeline
XB-BOX - 🎬 Wan 动画转视频

The engine under Wan Animate motion transfer — reference image + pose video in, sampled video out

XB_ToolBox/Wan
XB_WanCameraImageToVideo

Wan camera-control conditioning, with a start frame and tiled VAE encode

XB_ToolBox/Pipeline
XB-BOX - ⚡ Wan 编译设置

Torch.compile for Wan, as a settings bundle — and the reason AMD users should skip it

XB_ToolBox/Wan
XB_WanDancerCombo

One node that turns a music file into dance prompts and audio features for Wan Dancer

XB_ToolBox/Wan
XB_WanDancerVideo

Dance conditioning for Wan-Dancer — the official logic, tiled so it fits in VRAM

XB_ToolBox/Pipeline
XB_WanDanceSwitcher

The Wan Dancer switcher

XB_ToolBox/Wan
XB-BOX - Wan First/Last Frame2Video

Wan first/last frame animation that stays on rails

XB_ToolBox/Pipeline
XB_WanFunControlToVideo

One node for control-video conditioning

XB_ToolBox/Pipeline
XB_WanFunInpaintToVideo

Bookend-pinned editing for Wan 2.2

XB_ToolBox/Pipeline
XB_WanHuMoImageToVideo

Audio-driven human motion with Wan HuMo

XB_ToolBox/Pipeline
XB-BOX - Wan Image2Video

The node that turns your start image into Wan's I2V conditioning

XB_ToolBox/Pipeline
XB-BOX - Wan Infinite Relay Node

The relay node that chains Wan clips into a continuous video stream

XB_ToolBox/Pipeline
XB-BOX - 🆕 Wan 无限接力点 (New)

Relay_count turns N nodes into one

XB_ToolBox/Pipeline
XB-BOX - 🎵 InfiniteTalk 无限对口型总线

One bus for the whole talking-head pipeline

XB_ToolBox/Pipeline
XB-BOX - 🏃 InfiniteTalk 无限对口型接力点

A segment generator that knows where the audio ends

XB_ToolBox/Pipeline
XB-BOX - 🆕 InfiniteTalk 无限接力点 (全能)

Three ways to pick a segment's first frame

XB_ToolBox/Pipeline
XB-BOX - 🆕 InfiniteTalk 无限接力点 (多图)

Per-segment reference images from your input folder

XB_ToolBox/Pipeline
XB-BOX - 🆕 InfiniteTalk 无限接力点 (New)

Relay_count for talk-till-the-audio-ends

XB_ToolBox/Pipeline
XB-BOX - 🎵 语音转视频分块

One node for InfiniteTalk audio-to-video segments, single or dual speaker

XB_ToolBox/Pipeline
XB-BOX - 🎵 语音转视频分块(双人)

Dual-speaker InfiniteTalk chunks with per-person masks

XB_ToolBox/Pipeline
XB-BOX - 🎵 语音转视频分块(单人)

The no-surprises InfiniteTalk engine

XB_ToolBox/Pipeline
XB-BOX - Wan Param Bus

The Wan Param Bus that ends noodle-wire chaos

XB_ToolBox/Pipeline
XB_WanPhantomSubjectToVideo

Multi-reference conditioning with Phantom Subject

XB_ToolBox/Pipeline
XB-BOX - Wan Relay Node

Stitch a whole video from first-and-last frames

XB_ToolBox/Pipeline
XB-BOX - 🆕 SCAIL 总线 (New)

One control panel for a whole SCAIL relay chain

XB_ToolBox/Pipeline
XB-BOX - 🆕 SCAIL 无限接力点 (New)

Turn SCAIL into a five-minute video

XB_ToolBox/Pipeline
XB_WanSCAILToVideo

Plain SCAIL, one segment, no scaffolding

XB_ToolBox/Pipeline
XB_WanSCAILToVideoPro

SCAIL-2 with the relay bookkeeping built in

XB_ToolBox/Pipeline
XB_WanSoundImageToVideo

Make Wan video follow the audio

XB_ToolBox/Pipeline
XB_WanSoundImageToVideoExtend

Keep the audio going, extend the video

XB_ToolBox/Pipeline
XB_WanVaceToVideo

Region-aware Wan video editing

XB_ToolBox/Pipeline
XB-BOX - 🎵 YuE2 翻唱准备

YuE2 Cover Prep, explained

JZL/YuE2
XB-BOX - 🎵 YuE2 歌词

The eight section tags the model actually knows — and the lines it will sing at you

JZL/YuE2
XB-BOX - 🎵 YuE2 乐谱工作台

The YuE2 Score Workbench

JZL/YuE2
XB-BOX - 🎵 YuE2 风格提示

The YuE2 style string, built for you — and the one thing it can't do

JZL/YuE2
XB-BOX - 🎵 YuE2 转谱导入

What Transcription Intake fixes, and what it refuses to

JZL/YuE2
XB-BOX - Data Radar

A wire you can attach to see what your data weighs

XB_ToolBox/VRAM_Hacks
Readme
<div align="center"> 🌐 <a href="README.md">English</a> | <a href="README_zh.md">简体中文</a> </div> <br>

🧰 ComfyUI XB-BOX (XB_ToolBox)


🌟 Core Philosophy

The primary goal of XB_ToolBox is to help AI beginners new to ComfyUI quickly master workflows, making local deployment and execution simpler and more convenient. It is a comprehensive ComfyUI extension suite encompassing both front-end interaction and low-level memory scheduling.

  • Unified Parameters: Integrates frequently used but scattered parameter nodes, allowing you to set essential image and video parameters from a single hub.
  • Visual Learning: Provides visual chunking and parameter preview nodes, enabling beginners to intuitively understand the concepts of VAE/Encoder chunks, temporal slicing, and overlap effects.
  • UX Enhancements: Introduces a user-friendly "Mirror Clone Console" (Dashboard) that clones major operation nodes from complex workflows into a unified, clean, and highly efficient control panel.
  • Extreme VRAM Optimization: Through "Spatiotemporal Chunk Hijacking", "Dynamic VRAM Offloading for UNet/Checkpoints", and "Deep LiteGraph Customization", it allows consumer-grade GPUs to smoothly run massive 14B~22B 3D-DiT video models that would normally cause instant OOM (Out of Memory).

界面截图 界面截图


🗺️ What's inside — 141 nodes

| Area | Nodes | Highlights | |------|:---:|---| | 🎛️ Media parameters & I/O | 7 | Image / Video Params Master, Hailuo H3 params, Video Loader & Combine, Master Parameter | | 🧊 VRAM optimization | 25 | 17 official-node wrappers with tiered cleanup, Block Swap, SageAttention, GGUF loaders V1–V3 | | 🎬 Long-form video relays | 19 | InfiniteTalk / SCAIL / Wan Animate infinite relays, LTX-2.3 streamer, Video Merger, Storyboard Slicer, Digital-Human params | | 🎥 Wan node family | 24 | I2V, first/last frame, VACE, Fun, Camera, Phantom, HuMo, SCAIL, Dancer, Sound, InfiniteTalk — all with VAE tiling | | 🎵 Audio & speech | 15 | Audio Slicer ×4, CosyVoice3 ×10, Play Sound | | 🧠 Local LLM (XB-llama) | 17 | Loaders, instruct, parameters, prompt enhance / reverse, storyboard, MiniMax H3 presets | | 🖼️ Image presets | 4 | Image Prompt Preset / Pro, Qwen2.1 preset, ShenBi board | | 🖼️ Model-specific encoders | 3 | FLUX multi-image, Qwen image-edit (+ Plus) | | 🎨 Super routers | 2 | Bernini prompt router, K2 style router | | 🎵 YuE2 music | 5 | Style prompt, lyrics, score workbench, cover prep, transcription intake | | ✂️ Segmentation | 2 | ONNX human segmentation with a GPU (DirectML) path on AMD | | 💬 Comic | 3 | Dialogue parser, exact-coordinate text renderer, auto bubble renderer | | 🧰 Utilities & wiring | 11 | Batch loaders, image scale, list dispatcher, string merge, reference-any, status switch, label, MSR, name broadcasters | | 📊 Hardware dashboard | 1 | Live VRAM / RAM / CPU / GPU / disk panel |

Counts are read straight from NODE_CLASS_MAPPINGS in __init__.py.


✨ Core Nodes

1. 🎬 Media Parameters & Spatiotemporal Visualization

  • 🖼️ Image Parameters Master: Controls core image generation parameters (Width, Height, Batch Size, Strength) in one place. Enforces safe steps to prevent invalid inputs and automatically aligns dimensions when a fixed aspect ratio is selected.
  • 🎬 Video Parameters Master: Manages Width, Height, Frames, and FPS (Integer/Float). Features a geek-level "Auto-Manual Transmission Engine" that smoothly snaps to official pre-trained "Golden Buckets" (e.g., 480x832, 544x960) and strictly locks the physical frame count to the safe 1+8N rule. Dynamically displays video duration based on frame rate.
  • 🧊 Spatiotemporal Chunk Visualization: An exclusive dual-zone radar! The 2D left panel visualizes spatial image chunks and overlap values, while the 3D right panel uses a painter's algorithm to render temporal cylinder stacks. Saves beginners from the exhaustion of blind parameter tuning.
  • 🧊 Storyboard Image Slicer: Offers multiple modes to slice common 4-grid, 6-grid, 9-grid... images all at once and output multiple individual images. This makes image-to-video or start/end frame video generation much more convenient, and is highly effective for creating short dramas.
  • 📟 VRAM Calculator & Data Radar: Predicts VRAM footprint for WAN/LTX models under different quantizations (FP8/GGUF) and weighs tensor volumes with MB-level precision. Significantly reduces trial-and-error costs by helping users choose models fitting their VRAM capacity.

2. 🧊 VRAM Optimization

  • 🧊 Official Sampler / VAE Wrappers (12): call the official ComfyUI nodes verbatim — dynamic INPUT_TYPES, **kwargs passthrough, zero changes to the underlying algorithm — and inject a tiered VRAM cleanup (不做任何清理 / 卸载显存模型 — offload models to RAM, VRAM freed instantly / 卸载全量模型 — additionally drop model references and GC so both VRAM and RAM are released, models reload on next use) before each run. Covers KSampler, KSamplerAdvanced, SamplerCustom(+Advanced), VAEDecode(+Tiled / TiledImage), VAEEncode(+Tiled / TiledImage / ForInpaint), plus legacy ROCm-named aliases for old workflows (the 4 legacy sampler IDs, 3 legacy VAE decode IDs and 1 legacy VAE encode ID are auto-upgraded to the new nodes on load and no longer appear in the node menu).
  • ⚡ Auto un-tiled decode / encode: when the tile size covers the whole image, both tiled decoders switch to the official plain VAEDecode and the tiled image encoder switches to plain VAEEncode (the official 2D tiled paths run 3 tiled passes and average them; measured 40%~60% faster at 1024²).
  • ✂️ Block Swap (UNet / Checkpoint): forcibly parks the first N transformer blocks and the text/image embedding layers in system RAM, trading time for space.
  • 🧠 SageAttention Accelerator / Sage + Block Swap Duo: mount the attention-acceleration patch alone (auto / PyTorch SDPA / comfy-kitchen), or combined with physical block swapping in one node.
  • 📦 GGUF Model Loaders V1 / V2 / V3: all-in-one loaders stacking GGUF UNet + GGUF CLIP + LoRA (multi-slot) + VAE + Sage + Block Swap. V2 adds high/low-noise UNet pairs, V3 adds dual CLIP and dual VAE.

3. 🎛️ Workflow Deployment Assistants

  • 🎛️ Param Bus & Relay Nodes: an ultra-compact AnyType bus that holds every shared parameter (model, VAE, prompt, size, FPS, cleanup tier) plus a matching relay node that executes one segment and passes the accumulated result on. One node replaces a 10-node subgraph and sets your workflow free from "noodle wire" chaos.
  • 🧰 Small tools that save a workflow: list dispatcher (split one list into N outputs), reference-any (single-select switch over dynamic slots), node status switch (mute/bypass a target node from a boolean), string merge, name broadcasters, canvas label, Play Sound.
  • 📊 Mirror panel: the hardware dashboard clones the numbers you care about into one always-visible strip — see §13.

4. 🎬 Long-form Video Relays

A two-minute talking-head clip will OOM in a single pass on almost any GPU. These nodes split the job into reusable relay hops: a Param Bus holds every shared parameter (model, VAE, size, FPS, prompts, cleanup tier) and each Relay Node runs one segment (encode → sample → decode → trim → accumulate), then hands the accumulated result to the next hop.

  • 🎵 InfiniteTalk Param Bus / Relay — single speaker, dual speaker, multi-reference, and an all-in-one variant; the bus encodes the audio once and slices it per segment automatically.
  • 🏃 Wan Animate Param Bus / Relay — motion transfer from a driving video, same relay pattern.
  • 🆕 SCAIL Param Bus / Relay — SCAIL pose-driven video relay.
  • 🔄 LTX-2.3 Infinite Streamer — single-sample (8-step) long-form streamer with guide stitching.
  • 🎬 Video Merger / Storyboard Slicer / Digital-Human params — merge clips, slice grid storyboards into panels, and one-stop parameter + audio preparation for digital-human workflows.

5. 🎥 Wan Node Family (24)

Re-implementations of the Wan video nodes with VAE tiling built in (spatial + temporal chunking to flatten the decode peak): ImageToVideo, FirstLastFrameToVideo, VaceToVideo, FunControl / FunInpaint, Camera, PhantomSubject, HuMo, SoundImageToVideo (+ Extend), SCAIL (+ Pro), Dancer (+ Dance Switcher and Dancer Combo), InfiniteTalkToVideo (single / dual / legacy mode switch), Wan22ImageToVideoLatent.

These are graph members only — the Wan model / sampler / VAE / text-encode nodes come from ComfyUI-WanVideoWrapper. XB_WanAnimateToVideo and XB_WanCompileSettings are fully self-contained.

6. 🎵 Audio & Speech (15)

  • 🎵 Audio Slicer ×4 — waveform display with draggable cut handles: basic, V1 (single speaker), V2 (two speakers + gap), V3 (advanced, dual-track mixing).
  • 🔊 CosyVoice3 ×10 — model loader (auto-download + cache), zero-shot cloning, cross-lingual synthesis, instruct-style control, speaker-preset save / clone, voice conversion, multi-speaker dialog, audio crop. Presets are stored as CosyVoice-compatible .pt files, so you never re-upload the reference clip.
  • 🔊 Play Sound — play a notification when a workflow finishes.

7. 🧠 Local LLM Prompt Suite (XB-llama, 17)

Local LLM/VLM inference for prompt work: model loader (+ Pro, which folds every inference parameter into one panel), instruct, reasoning parameters, unload / clean-state, bbox extraction, code-block unpacking, prompt enhancer, prompt reverse (a five-in-one node: loader + parameters + API settings + preset + instruct), storyboard enhancer / instruct / processor (+ Pro), MiniMax H3 prompt presets (fl2va, ref2va) and the role/scene dispatcher.

Both a local model and an online API backend are supported; API keys live in the ComfyUI user directory and are never written into the workflow.

8. 🖼️ Image Prompt Presets (4)

  • 🖼️ Image Prompt Preset — one node replacing "prompt box + aspect-ratio node + official empty latent", with step-locked width/height and ratio linking.
  • 🖼️ Image Prompt Preset Pro — the same, plus built-in LLM prompt enhancement / reverse.
  • 🖼️ Qwen2.1 Prompt Preset — preset + element panel, wired to the official Load CLIP + Generate Text chain.
  • 🖌️ ShenBi — an input-less paint board (create / edit modes) that outputs a strictly on-resolution image.

    The preview window reads files from this server's input dir (boards live in input/xb_shenbi/). A workflow only stores the file name, so after copying it to another machine (local ↔ cloud) the window falls back to a blank canvas and explains why in the badge / browser console. The window itself needs a recent ComfyUI frontend (DOM widgets).

9. 🖼️ Model-specific Encoders (3)

  • FLUX Multi-Image Encode — resize → VAE encode → ReferenceLatent chained, with image slots auto-expanding up to 9.
  • Qwen Image-Edit Encode (+ Plus) — ports of the official TextEncodeQwenImageEdit / ...Plus, dynamic image slots up to 9.

10. 🎨 Super Routers (2)

Bernini Prompt Router and K2 Style Router rewrite heavy multi-switch subgraphs (Fast Muter + Any Switch chains) as a single native node, so the front-end cost of refreshing and dragging drops from O(n) hidden nodes to O(1).

11. ✂️ Segmentation & Comic (5)

  • ✂️ Human Segmentation — ONNX u2net (RMBG-1.4) model loader plus a mask / cutout node, running on the GPU through DirectML on AMD.
  • 💬 Comic ×3 — parse dialogue out of a script, render text at exact coordinates, or auto-detect white speech bubbles and fill them in.

12. 🧰 Utilities & Wiring (11)

Batch folder / batch images (muted inputs are skipped), batch image scale (N-in / N-out), list dispatcher, string merge (with a per-segment newline option), reference-any, node status switch, canvas label, MSR multi-image → frame sequence, and UNet / CLIP name broadcasters.

13. 📊 Hardware Dashboard (1)

XB 📊 Memory Visualization — a live panel for VRAM / RAM / CPU / GPU / disk / pagefile, with a per-model VRAM breakdown, an aimdo page-residency heatmap, history graphs, themes, and a docked mini-bar that pins to the ComfyUI top bar. Full guide: docs/XB_MemoryVisualization_manual.md.

14. 🎵 YuE2 Music

YuE2 is a white-box song generator: it plans an editable ABC score from style + lyrics, then renders audio from that score. These nodes bring the official skill's symbolic-layer workflow into ComfyUI (the ABC parser is an in-node port of the official bounded-dialect implementation, standard library only, nothing leaves your machine).

  • 🎵 YuE2 Style Prompt: Builds the official six-part style string (Language / Genre / Vocal / Tempo / Instruments / Musical Characteristics) with 13 one-click recipes and a genre library. Also exposes max_duration and seed, so the planner and the renderer share one seed for reproducible plans. Structure, key and meter stay out of style — they belong to the lyrics tags and the score.
  • 🎵 YuE2 Lyrics: Section tags with a picker (the official eight-tag vocabulary + community spellings + your own [custom] tags), a one-click instrumental skeleton, and singability checks (implementation notes leaking into lyrics, round-bracket role labels that get sung, empty sections, per-line syllable counts, and whether a tag is official).
  • 🎵 YuE2 Score Workbench: Inspect / strip chords / move the vocal melody to Ins / use an edited score, with the official rule set (grid, Z2–Z4 compression, chord vocabulary, ties). Emits the processed score plus a readable report whose first line is the mode suggestion (full vs melody). Note the score carries no instruments — set those in the style node. Option names stay short; every explanation lives in the widget tooltip.
  • 🎵 YuE2 Transcription Intake: Reads SheetSage2 output (score.abc / events.json) from a folder or file, applies the mechanical normalizations from the official repair table (Markdown fences, BOM, curly quotes, M:C, Q:81, K:B minor), validates against the native dialect and emits a usable score. Semantic repairs (unrolling repeats, quantizing tuplets, adding rests) are deliberately left to you, with the official validator's own message shown.
  • 🎵 YuE2 Cover Prep: Builds cover conditions — which voice to keep x how to handle the harmony — and the harmony choice literally sets the downstream mode (chord-free → melody, a true cover; keep chords → full, closer to a re-record). The report carries the before/after event comparison and the official cover flow. A cover only supplies a symbolic melody; it does not preserve the source singer's identity or waveform.

Requires a ComfyUI build that ships the native YuE2GenerateABC / YuE2GenerateMusic nodes (bring them in with the official text-to-music workflow). The nodes here only emit STRING/FLOAT/INT, so they load on any build.

Where each official capability lives (so nothing is duplicated or missing):

| Official capability | Where it is | |---|---| | Planning/rendering parameters — mode, max_abc_tokens, temperature, top_p, top_k, repetition_penalty, penalty_window, cfg_scale | The official YuE2GenerateABC / YuE2GenerateMusic nodes. This package only feeds style / max_duration / seed into them. | | Style, lyrics, section tags | 🎵 YuE2 Style Prompt / 🎵 YuE2 Lyrics (the official eight-tag vocabulary + community spellings + custom tags) | | ABC inspect / strip chords / move voice / event compile / difference review | 🎵 YuE2 Score Workbench (a port of the official bounded-dialect implementation) | | Cover conditions (chord-free → melody, keep chords → full, voice selection) | 🎵 YuE2 Cover Prep | | Decoder choice (YuE2-Vae for listening, YuE2-Vae-legacy for benchmark reproduction) | Determined by the VAE you load; re-decoding cached latents needs no new generation. | | Audio→score transcription (SheetSage2) itself | Not here — it needs a separate Python environment (Python 3.10/3.11 + torch 2.8/cu126). Use 🎵 YuE2 Transcription Intake to bring its score.abc / events.json back into the graph. | | MERT2 features, yue2 batch, the listen.py HTML player, SongBench scoring | Not here — they are delivery/evaluation tooling rather than graph nodes. |

Handy official behaviours the nodes surface for you: off mode is automatic when the ABC input is empty, a truncated score fails structure validation, mode=melody rejects any chord symbol, and the planner and renderer can receive different lyrics (section-tags-only for planning, real words for singing).

Instrumental needs three things, not just a style setting. The official instrumental entry point (a) prefixes Instrumental, and (b) forces four suppression conditions — no vocals, no singing, no choir, no spoken words. The style node now does exactly that when you pick 纯音乐. But style alone cannot remove singing: the score's Vocal voice must have no sounding notes (Score Workbench → "Vocal to Ins"), and the lyrics must be empty or section tags only (Lyrics node → "纯音乐骨架"). The official docs still say a file check cannot prove the absence of vocal leakage — confirm by listening.


📦 Installation

Method 1: ComfyUI Manager (Recommended) Search for XB_ToolBox in the ComfyUI Manager and click install.

Method 2: Manual Git Clone Navigate to your ComfyUI custom_nodes directory and run:

git clone https://github.com/wjluoxiao/XB_ToolBox.git

(Note: the core architecture relies purely on the native ComfyUI ecosystem — elegant Python/JS, no patched libraries. Some node groups do import extra packages; see below.)

🔧 Dependencies

Most nodes are dependency-free, but a few groups import extra packages. Install them all with:

pip install -r requirements.txt

| Node group | Packages | |---|---| | Audio slicer, video I/O | av, imageio-ffmpeg | | Comic text detection | easyocr, opencv-python | | Human segmentation | onnxruntime (on AMD use onnxruntime-directml) | | Hardware dashboard | psutil, pythonnet (bundles LibreHardwareMonitorLib.dll) | | Local LLM (XB-llama) | diskcache, gguf, and llama-cpp-python — install that one manually to match your GPU | | CosyVoice3 | modelscope, huggingface_hub, transformers, librosa, soundfile, HyperPyYAML, omegaconf, hydra-core, conformer, openai-whisper, tiktoken, scipy, einops, … | | Audio waveform endpoint | aiohttp |

Optional companion extensions — a few groups expect these to be installed:

| Feature | Needs | |---|---| | Wan nodes & relay pipelines | ComfyUI-WanVideoWrapper — supplies the Wan model loader / sampler / VAE / text-encode nodes | | GGUF model loaders | ComfyUI-GGUF | | YuE2 preset nodes | A ComfyUI build that ships YuE2GenerateABC / YuE2GenerateMusic |

Nodes that are fully self-contained say so in their own docstring.


🗃️ Repository layout

One feature per folder; the root only keeps the assembly entry points:

XB_ToolBox/
├─ __init__.py               node registration + route wiring
├─ prestartup_script.py     startup cleanup (ComfyUI convention: must stay at root)
├─ nodes/                    node implementations
│  ├─ core/                 parameter bus / relay / label / string / switch / accel
│  ├─ image/                prompt presets / encoders / scale / segmentation / comic
│  ├─ video/                Wan / LTX / Hailuo / digital human / video I-O
│  ├─ audio/                audio slicer / YuE2 / play sound
│  ├─ llm/                  XB-llama / GGUF loaders
│  ├─ hardware/             VRAM & memory monitoring
│  └─ cosyvoice3/           CosyVoice3 node package
├─ libs/                    node-free logic (xb_paths / xb_presets / xb_gallery / xb_skills / yue2_* …)
├─ jzl_workflows/           "JZL Workflows" (data + routes + icon/style)
├─ web/                     frontend assets (formerly js/)
├─ locales/ docs/ tests/    i18n / docs / tests
├─ tools/                   one-off dev scripts (not used at runtime)
├─ support_llama/ xb_skill/ video_formats/    data folders
└─ LibreHardwareMonitorLib.dll

🗂️ JZL Workflows (sidebar entry)

Once XB_ToolBox is installed, ComfyUI's sidebar gains a "机智罗 / JZL Workflows" entry, sitting right above the built-in Templates tab. Clicking it opens a centered dialog just like the official Templates browser — category list on the left, search on top, thumbnail cards, and a card click opens the workflow in a new workflow tab (Esc / backdrop click / ✕ closes it). No ComfyUI frontend files are patched, so frontend upgrades won't break it.

Everything for this feature lives in jzl_workflows/:

XB_ToolBox/
  jzl_workflows/
    __init__.py        registration entry (root __init__.py calls it)
    routes.py          routes: index / workflow / preview / icon+css assets
    data/
      index.json       English index
      index.zh.json    Chinese index (same shape as comfyui-workflow-templates)
      anima_t2i_basic.json      the workflow itself
      anima_t2i_basic-1.webp    preview image; a placeholder card is shown if missing
    web/
      xb_workflow_gallery.css
      xb_icon_dark.png / xb_icon_light.png
  web/jzl_workflows/
    xb_workflow_gallery.js       panel script (must stay under web/ to be loaded)

To add a workflow: drop the JSON into jzl_workflows/data/ (ASCII file name), add an entry to both index files (name must match the file name), and put the preview next to it as <name>-1.webp.


🙏 Credits

XB_ToolBox stands on the shoulders of the ComfyUI ecosystem. Several nodes are ports, wrappers or adaptations — thanks to their authors:

  • VideoHelperSuite (VHS) — the video loader/combine and audio-slicing logic derive from VHS.
  • rgthree-comfy — XB_ReferenceAny fuses Fast Muter + Any Switch.
  • ComfyUI-KJNodes — XB_StringMerge is reworked from JoinStringMulti; the SageAttention node keeps the original KJNodes design.
  • ComfyUI-Custom-Scripts (pysssss) — XB_PlaySound is an independent re-implementation.
  • ComfyUI-GGUF (city96) — the GGUF loaders wrap its official nodes.
  • ComfyUI-WanVideoWrapper (kijai) — companion extension for the Wan family.
  • ComfyUI-MemoryVisualization (kijai) — the hardware panel started from this project and has since grown its own themes, docking and metrics.
  • ComfyUI-llama-cpp_vlm — the XB-llama nodes are adapted from it.
  • XB_KSampler and friends call the official ComfyUI nodes verbatim and only add cache cleanup.

📝 License This project is open-sourced under the Apache-2.0 License. It guarantees the freedom to share while protecting the patent defense rights of the core architecture code.