Encontre o que você precisa para seu fluxo de trabalho.

Wan 2.1 ComfyUI Repackaged

Repackaged model files for ComfyUI.

LTX 2

ltx-2-community-license-agreement

FLUX.2 Klein Base 4B

The FLUX.2 [klein] model family are our fastest image models to date. FLUX.2 [klein] unifies generation and editing in a single compact architecture, delivering state-of-the-art quality with end-to-end inference in...

SDXL Turbo

sai-nc-community

XTTS v2

coqui-public-model-license

XTTS v2

coqui-public-model-license

A1111 To Comfyui workflow (My version) — v2.2

Removed nodes that where redundant, as well getting rid of the styles node. Got rid of nodes from comfy-image-saver, Due to problems. Not Comfyui-image-saver. Two different plugins.

FastVideo

A unified inference and post-training framework for accelerated video generation.

ComfyUI-Easy-Use

In order to make it easier to use the ComfyUI, I have made some optimizations and integrations to some commonly used nodes.

ComfyUI-HunyuanVideoWrapper

ComfyUI-RMBG

A ComfyUI custom node designed for advanced image background removal and object, face, clothes, and fashion segmentation, utilizing multiple models including RMBG-2.0, VRMBG-3.0, FeyNoBg, Finegrain Box Segmenter,...

sd-ppp

A Photoshop AI plugin

comfyui-mixlab-nodes

x-flux-comfyui

ComfyUI-segment-anything-2

ComfyUI nodes to use segment-anything-2

Combined character sheet

Generate separate headshot and full-body character panels and arrange them into one sheet. This tool runs two Gemini image generations before stitching their images; it does not accept pre-generated panels. For a first run, upload a character reference in Source image, retain both default prompts, the headshot panel's 9:16 ratio, the full-body panel's 16:9 ratio and both 2K resolutions, then run. The prompts request four close-up views and four full-body views on white backgrounds. To generate without a reference, bypass Source image and describe the same character in both prompts. Leaving the upload empty alone does not select text-only generation. Each panel has its own prompt, aspect ratio, resolution, model, seed, response format and system instructions. Keep character and style descriptions aligned across the prompts. Aspect ratio controls each complete generated panel, not the individual views within it. Auto follows the reference image's ratio when supplied; choose an explicit ratio without a reference when output shape matters. Resolution offers 1K, 2K and 4K; 2K and 4K use Gemini's native upscaler. Headshot position places the headshot panel relative to the full-body panel: right, left, above (up) or below (down). The default places it on the right. Match panel size resizes panels along the joining edge; disabling it retains their generated sizes. Panel spacing sets the gap in pixels, from 0 to 1024 in increments of 2. Spacing color affects that gap when spacing is greater than zero. The default uses size matching with no gap. System instructions set each provider request's foundational behavior; retain an explicit image-generation instruction. IMAGE+TEXT can request accompanying provider text, but this tool returns only the stitched image. Seed -1 requests a new random seed for each panel each run. Fixed seeds provide best-effort repeatability rather than deterministic results. Requested view counts, poses and consistent character identity are generation instructions rather than guarantees. The two generations may differ in appearance, and size matching adjusts the layout rather than reconciling their content.

Seedance text to video

Generate one short video from a scene description with Seedance, with optional generated audio on Seedance 1.5 Pro. No uploaded media is required. For a first run, keep Seedance 1.5 Pro selected and use the example prompt or describe your own subject, setting, action, and camera movement. The defaults request a five-second, 720p, widescreen video with generated audio. Include dialogue or sound intent in the prompt when audio is enabled, and keep the action and dialogue appropriate for the chosen duration. Choose a duration from 3–12 seconds, a resolution of 480p, 720p, or 1080p, and the aspect ratio for your intended frame shape. Model lets you select the listed Seedance 1.0 alternatives; those models ignore Generate audio. With Seedance 1.5 Pro, turn Generate audio off for a silent result. Fixed camera adds a stationary-camera instruction to the prompt but does not guarantee camera lock. Remove conflicting camera movement instructions when enabling it. AI watermark requests an “AI generated” watermark. Seed -1 randomizes each run; a nonnegative seed controls the generation's random input. The generated video is returned through Sweet Tea's video output transport with automatic format and codec selection by default. Generation sends your prompt to the ByteDance provider and uses provider-priced billing.

Poster remake and image edit

Create a new poster from a reference image, apply direct image edits, or generate an image from text using Gemini. For a first run, upload a reference poster, leave Direct prompt mode off, and replace the example Prompt with your new title, supporting text, subject, and desired content. Leave the Reference image stage enabled when using an upload. The default Poster instructions ask Gemini to borrow the reference's style, colors, and typography while changing people's faces, clothing, and poses. They request a large, mainly white artistic title that avoids the face and omit reference watermarks and tiny text. These are creative directions, not guarantees of exact layout, lettering, or identity preservation. Turn Direct prompt mode on to send your brief without Poster instructions. With a reference enabled, describe the edits to apply. For text-only generation, disable the Reference image stage and describe the image you want; Direct prompt mode avoids the default poster instructions that refer to an uploaded picture. Generation instructions still apply in both modes. Auto aspect ratio follows the reference when present. Choose an explicit ratio when you need a particular format, especially without a reference. Resolution defaults to 2K; the source uses Gemini's native upscaler for 2K and 4K. Model selects the Gemini image model for all modes. Use seed -1 for a new random seed each run; a fixed seed requests best-effort consistency, but identical results are not guaranteed. Adjust Poster instructions under advanced controls to change the reusable poster direction. They are prepended to your brief only when Direct prompt mode is off. Generation instructions control the shared foundational direction sent to Gemini. This Pipe returns generated images through its preview output; it does not include the model's separate text response or a workflow file-saving stage. Prompts and enabled reference images are sent to Gemini through a provider-priced API call.

Stable Diffusion XL Refiner 1.0

SDXL consists of an ensemble of experts pipeline for latent diffusion: In a first step, the base model (available here: is used to generate (noisy) latents, which are then further processed with a refinement model...