ComfyUI Custom Extensions

ComfyUI custom node - universal model quantizer. Converts diffusion models (safetensors) to NVFP4 / FP8 (e4m3fn) / GGUF in place, with a per-architecture blacklist that protects structure-sensitive layers so the result loads and renders correctly. Auto (Universal) mode works on any architecture.

Local metadata backend for ComfyUI Prompt Workbench: offline safetensors trigger scanning, editable character/outfit groups and output organization. Optional explicit online scans query Civitai using SHA-256 hashes. No models or personal database bundled; no standalone panel.

Sidebar for prompt composition, field templates, LoRA character/outfit selections, saved batch presets and pause/resume submission. Chinese/English UI and local backups. Install ComfyUI LoRA Trigger Helper for metadata scanning/editing and character output organization.

ComfyUI nodes for PhotoDoodle model.

Nodes:Upscale Video Tensorrt

A local MiniMax H3 creation workspace for ComfyUI with visual prompt authoring, Cast, references, timeline direction, LoRAs, PreStage, sampling, and output tools.

A custom node for ComfyUI that performs color correction on clothing in face-swapped images. This node helps maintain the original clothing color when using face swap tools, addressing common color shifts that occur during the face swap process.

A ComfyUI Custom Node to backup workflows and their required models locally.

EmbeddingsNameLoader, EmbendingList

A set of nodes for rapidfiring the half backed latents, cleaning up obvious bad generations and automatically queueing the rest to fully generate.

ComfyUI node for creating some Turtle Graphic demos.

ComfyUI custom nodes for integrating multiple external APIs including Gemini, OpenAI, Replicate, and ElevenLabs directly into local workflows. (Description by CC)

Dropdown-based prompt builder for ComfyUI. Create characters, scenes, and styles without writing prompts. Designed for beginners and fast iteration.

ComfyUI custom nodes for converting prompt text between common anime image-generation formats.

ComfyUI WorkspaceKit (WK): a modular workspace, sidebar, and panel system for organizing workflows, nodes, favorites, templates, and canvas groups in ComfyUI.

Nodes:ZC DrawShape Node

Provide a simple interface to simplify prompt writing.

A ComfyUI custom node that patches the attention mechanism inside a VAE's encoder and decoder to use a specific implementation, regardless of what ComfyUI auto-selected at startup.

A LLM API key management solution for ComfyUI with automatic load balancing support.

You'll get a new node Latent Size Selector, you can pick the x and y sizes from a list.

DeepSeek-powered Illustrious prompt generator nodes for ComfyUI.

This project allows you to adjust SDXL's two text encoder's strengths individually for clip_g (ViT-bigG) and clip_l (CLIP-ViT-L) within ComfyUI. (And other adjustments)

One-stop ComfyUI canvas organizer: align, resize, flow-arrange, recolor, snap-drag and group nodes — 14 alignment modes, 14 palettes, EN/CN UI

A simple ComfyUI nodes that integrates [a/PiAPI faceswap](https://piapi.ai/faceswap-api) service into ComfyUI. This can be helpful if you're trying to create a workflow that includes faceswap for commercial usage.

ComfyUI custom node suite integrating multiple AI image generation APIs (Gemini, Flux.1 Kontext Pro) with unified authentication via ComfyUI API Key and real-time cost tracking. (Description by CC)

为 ComfyUI 引入 Google NanoBanana Gemini 文生图/图生图能力的自定义节点,提供批量生成、并发调度与余额监控等高级特性。

为 ComfyUI 引入 Google NanoBanana Gemini 文生图/图生图能力的自定义节点,提供批量生成、并发调度与余额监控等高级特性。

ComfyUI custom nodes for Hunyuan3D - Complete 3D generation suite

ComfyUI nodes for integrating Ollama local LLM models with image or text generation workflows.

This extension can send HTTP Requests. You can request image generation to StableDiffusion3 and post images to X (Twitter) and Discord.

A custom ComfyUI node that generates ready-to-use image prompts with cloud or local multimodal language models.

RAFT-based optical flow nodes for ComfyUI with robust batch/index handling, flow save/load, and inpaint crop/stitch support.

Nodes:Blender viewport color, Blender Viewport depth

基于 Qwen3 的本地翻译节点

Warp Unfake.py for ComfyUI

Word-level audio-text alignment with multiple engines: LIS token matching, stable-ts cross-attention, and Qwen3 forced alignment. Supports faster-whisper, UVR5 vocal separation, and video subtitle overlay rendering.

A ComfyUI custom node for generating dynamic scrolling lyrics videos with smooth animations. Supports audio-to-text transcription using OpenAI Whisper or importing existing SRT subtitles.

ComfyUI custom nodes for LingBot-Video-Dense-1.3B — text-to-video and text+image-to-video on consumer GPUs.

A single custom node, Load Image (Smart), that addresses several shortcomings of ComfyUI's built-in image loaders. It includes: Base folder selector – choose between output and input; subfolder selector – lists single-level subfolders of the selected base folder; videos filtered out – the image list only shows files whose mimetype is an image; selection survives reloads – unlike the built-in Load Image (Outputs) node, the saved selection is preserved when a workflow is reloaded or refreshed.

Pillar is an extension plugin for ComfyUI, providing the ability to call local distributed services for ComfyUI. Currently, it integrates the llama-joycaption-beta-one-hf-llava model and offers a distributed deployment solution to solve the problem that the model occupies too many resources and affects the image generation efficiency.

Compare two ComfyUI videos with a synchronized swipe slider. Review upscaling, reference matching, and A/B results, then record the comparison to Save Video.

This custom node helps to generate cation for images for lora training.

ComfyUI V3 custom nodes for exchanging images and workflow parameters with Photoshop.

Add data chain operation between blender and comfyui. Provides ComfyUIInputs node for blender to fetching data from ComfyUI, and Blender Inputs/Outputs node for ComfyUI to fetching/sending data with blender.

Universal prompt optimizer for any ComfyUI model, with R2V bridges for both MiniMax H3 Director and TimelineDirector.

A lightweight custom node for **ComfyUI** that lists and batches files from a folder with filtering and sorting options.

Nodes: LiamLibLoadImage, LiamLibImageToGray, LiamLibSaveImg, LiamLibFillImage, PreviewReliefImage, GetBetterDepthImage, LiamLibSaveText

Stop wasting hours tweaking prompt. This extension allows you to modularize your prompts into variables and assemble them using a powerful template-based approach.

Dual-source image gallery for ComfyUI: browse, search and filter ComfyUI & Stable Diffusion WebUI outputs, with artist gallery, artist database, favorites and advanced tag search.

This is a simple node for connecting images. For pictures of the same size, users can choose to fill in vertical in the parameter to connect the pictures vertically or fill in horizontal to connect the pictures horizontally.

A chat assistant that lives inside ComfyUI's web UI — picks workflows, generates new ones from scratch, debugs broken graphs, all from natural language.

This ComfyUI custom node enhances the standard CLIP text encoding functionality by integrating ethnicity and hairstyle selection into the positive prompt. Designed to work seamlessly with ComfyUI, this node allows you to influence the conditioning process by dynamically appending descriptive modifiers. Users can choose a specific ethnicity or hairstyle, or opt for a 'random' selection that picks an option from a predefined CSV list.

Predicts Physical FPS (PhyFPS) from video using the Visual Chronometer model from the Pulse of Motion paper. Includes SDPA-optimized attention, device selection, progress tracking, and auto-downloads the model from HuggingFace.

ComfyUI nodes for seamless tiling and Akatz workflow utilities.

UI extensions: Control Multiple Samplers, Project Settings. Nodes: AK Base, AK Index Multiple, IsOneOfGroupsActive, RepeatGroupState group enable/disable like rg but without connection, Pipe, Getter, Setter.

Utility tools. Index Multiple, CLIP Multiple, nodes to enable diable groups.

Basic Color Nodes for ComfyUI

Implementation of Normalized Attention Guidance: Universal Negative Guidance for Diffusion Models for ComfyUI.

Custom nodes pack for ComfyUI

Several utility nodes for ComfyUI: LoRA management, image annotation, project organization, sampling utilities, system integration.

A ComfyUI Note node with HTML capabilities. Create notes, UI's and ReadMe's directly inside the workflow using rich HTML formatting.

A single-node ComfyUI module for Invertible Image Rescaling (IRN) that includes: automatic model download and discovery, BasicSR-based IRN_x2/x3/x4 loading, device/precision control, low-VRAM tiling, safe handling of non-divisible sizes, alpha channel preservation, and built-in bicubic/area fallback.

ComfyUI custom nodes for selecting media ranges, editing and comparing audio, and previewing audio and video.

ComfyUI custom node integrating directly with the Black Forest Labs FLUX 3 Video API. Check us out a demo of this node on YouTube https://youtu.be/4MDFL5CmF24

A set of nodes for generation of simple 2D graphics ComfyUI. These are intended for guiding image generation, for instance using controlnets.

Workflow toolkit for ComfyUI. Includes BiRefNet background removal, dynamic prompt/image variable system for batch generation, chroma key tools, and more utilities.

Image metrics nodes for ComfyUI

ComfyUI custom nodes for sharding SD/SDXL/Flux checkpoints across multiple GPUs.

Nodes:Cascade_Resizer

Nodes: Easy Resolution Picker, Save Diffusion Model, Load Checkpoint BNB On the fly, Load UNET BNB On the fly

ComfyUI custom nodes for GPT-Image-2 — OpenAI's latest image generation model. Generate and edit stunning AI images directly inside ComfyUI using the muapi.ai API.

Real-time ComfyUI Hardware Monitor with native AMD ROCm and All GPU/APU NVIDIA support and Universal Prompt Translator node for multilingual AI workflows, AMD ROCm utilities, optimization, and workflow enhance

custom nodes integrating Supertone's Supertonic-3 — a lightning-fast, on-device, multilingual Text-to-Speech system running natively via ONNX Runtime.

Super Fast LLM's llama3.1-405B,70B,8B and more

Training-free Hunyuan3D shape-generation acceleration: skip DiT steps and forecast the flow-matching velocity instead (HiCache Hermite / HiCache++ DMD, via hicache-pp). Pairs with kijai/ComfyUI-Hunyuan3DWrapper.

Training-free TRELLIS image-to-3D acceleration: forecast the flow-matching velocity on skipped DiT steps (HiCache Hermite / HiCache++ DMD, via hicache-pp) across both the sparse-structure and SLaT stages. ~2x faster, near-lossless. Drop between the TRELLIS loader and sampler.

Advanced model merging and enhancement nodes for ComfyUI with sophisticated techniques for blending, enhancing, and manipulating Stable Diffusion models

Show Text; Filename Prefix By Date; Positive Only (CLIP Encode Prompt); ...

ComfyUI Qwen 3.5 Prompting nodes (captionate image, improve prompt, etc)

ComfyUI node that lets the user pick a folder on their own machine and, on every workflow run, uploads exactly one file from it to the server.

Bridge VHS_VideoCombine video outputs to standard image payloads for serverless runners (RunPod, Modal). Enables automatic S3 uploads of MP4/WebP files in worker-comfyui deployments.

ComfyUI学习中心插件，提供交互式学习体验，帮助用户从零开始学习ComfyUI。

这是一个用于ComfyUI的生命游戏自定义节点，支持动态运行和实时控制。 A Conway's Game of Life custom node for ComfyUI with dynamic execution and real-time control

Advanced model merging and enhancement nodes for ComfyUI

Clean, practical node pack for LoRA management and image enhancement with LoRA Loader, LoRA Pro, LoRA Stack, and Detail Enhance nodes compatible with SD 1.5, SDXL, Flux, and HiDream. (Description by CC)

Design-focused node suite for ComfyUI. Convert raster to vector with a built-in PNG to SVG pipeline — bring true vectorization to your generative workflows and stop exporting pixel-locked assets. Includes precision utilities for grid systems, construction lines, edge detection, masking, image transforms, cropping, JSON tools, and flow control. Built for logo design, brand asset pipelines, and anyone who needs clean scalable output from their design based workflows. API tokens for vector nodes available at [augment studio](https://augmentstudio.app/pricing)

A simple ComfyUI node that allows you to use recent output images as input in your workflows.

A custom nodes package for ComfyUI that enhances workflow flexibility by providing specialized nodes for saving and loading intermediate data (encoded prompts and sampled latents) in multiple formats. This package leverages helper classes for file I/O, supports gzip compression for efficient storage, and integrates progress feedback via a progress bar to improve user experience during long operations.

Split a multiline prompt into up to 12 numbered STRING outputs using a user-defined delimiter. Useful for multi-segment video workflows.

Comfyui custom nodes to handle video concatenation with FFmpeg

Extend MiniMax H3 video in both spatial and temporal dimension

This custom node for ComfyUI provides a wrapper for Intel's Open Image Denoise (OIDN) library, allowing you to denoise images directly within your ComfyUI workflow.

Tiny custom node that adds drag & drop file upload to any hosted ComfyUI where you don't have shell access.

Strip metadata & workflow info from ComfyUI output images/videos (PNG/JPEG/MP4/WebM/GIF/WebP, lossless). 无损去除图片/视频中的提示词、工作流与EXIF信息。

This is custom node for audio transcribe with SRT.

Interactive editor for drawing animated, keyframed masks on a video/image batch in ComfyUI — circle, rectangle and bézier shapes with automatic keyframing, feathering and invert.

A set of ComfyUI nodes designed to enhance your workflow with realistic filename generation and keyword generation.

This repository contains a simple custom node for ComfyUI that implements familiar PS-style blend modes using PyTorch. The PSBlendNode allows you to blend two images together using a variety of blend modes and an opacity parameter.