Go to OpenArt main site
Upload workflow

FLUX All-in-One Workflow(Controlnet/Lora/Ipadapter/Upscale)

5.0

1 reviews
7
2.3K
814
7
Description

小红书:362973549    

xiaohongshu:362973549

此工作流全面支持FLUX模型的各类应用。

This workflow fully supports all types of applications for FLUX models.

支持Controlnet/LORA/Ipadapter/高清修复(潜空间、SD放大)

Controlnet/LORA/Ipadapter/HD restoration support (subspace, SD zoom)

文生图Vincennes:

添加画面描述和画面细节描述,越丰富越好并选择合适的分辨率比例,最终图片分辨率会乘1.5倍。

Add a description of the picture and the details of the picture description, the richer the better and choose the appropriate resolution ratio, the final image resolution will be multiplied by 1.5 .

(可选)选择lora,并在两个描述的最后都加上lora的触发词,不需要那么多lora可以关闭开关。

(Optional) Select lora and add lora trigger word at the end of both descriptions, you can turn off the switch if you don't need that much lora.

(可选)可以先关闭丰富细节和细节再优化开关,等第一张图效果满意再打开会快些。

(Optional) You can turn off the rich detail and detail re-optimization switches first, and then turn them on again when you are satisfied with the results of the first image, it will be faster.

(可选)可结合ipadapter,上传参考图来协助生成.

(Optional) can be combined with ipadapter, upload a reference picture to assist in generating.

LORA开关不作介绍,丰富细节和细节再优化开关,分别是利用潜在放大和SD放大方式来提升细节。

The LORA switch is not introduced, and the Enrich Details and Re-optimize Details switches are used to enhance details by using potential zoom and SD zoom, respectively.

图生图Diagram generation::

在文生图的基础上,再开启图生图和Canny/Depth/HED(线稿图)的开关,加载图片,设置分辨率,输入提示词。

On top of the text-generated map, turn on the graph-generated map and Canny/Depth/HED (line drawing) switches, load the image, set the resolution, and enter the cue word.

(可选)可结合ipadapter,来生成.

(Optional) can be combined with ipadapter to generate

(可选)选择lora,并在两个描述的最后都加上lora的触发词,不需要那么多lora可以关闭开关。

(Optional) Select lora and add lora trigger word at the end of both descriptions, don't need that much lora can turn off the switch.

图片示例介绍(Picture Example Introduction):

Black Myth Wukong style repainted(黑神话悟空风格转绘)

You need to upgrade to the latest Comfyui, and the nodes that appear can basically be installed directly in the manager(需要升级到最新的Comfyui,出现的节点基本都可以在manger里面直接安装。)

模型和节点Models and nodes:

ipadapter:

https://huggingface.co/XLabs-AI/flux-ip-adapter/tree/main

flux:

https://huggingface.co/black-forest-labs/FLUX.1-dev/tree/main

Xlabs:

https://github.com/XLabs-AI/x-flux

controlnet:

https://huggingface.co/XLabs-AI/flux-controlnet-collections/tree/main

lora(wukong):

https://civitai.com/models/681691

https://drive.google.com/drive/folders/1XOqGPZQccJGAMTITWT4WFfSXQf_xCDSM?usp=sharing

Xlabs-lora:

https://huggingface.co/XLabs-AI/flux-lora-collection/tree/main

缺少的模型都可以在huggingface上下载。

The missing models are all available for download on huggingface.

缺少的节点部分可以通过管理器安装,无法安装请移至相关节点的github主页学习安装。

The missing node part can be installed by the manager, if you can't install it, please move to the github page of the relevant node to learn how to install it.


Discussion

(No comments yet)

Loading...

Author

3
1.6K
14
5.4K
    20240902.json (73.8 kB)

No reviews yet

  • - latest (a month ago)

  • - v20240828-085440

  • - v20240828-084518

Primitive Nodes (26)

ApplyAdvancedFluxIPAdapter (1)

ApplyFluxControlNet (3)

CLIPTextEncodeFlux (2)

Fast Groups Bypasser (rgthree) (1)

FluxLoraLoader (7)

LoadFluxControlNet (3)

LoadFluxIPAdapter (1)

ModelSamplingFlux (1)

Note (2)

Reroute (4)

XlabsSampler (1)

Custom Nodes (41)

AlekPet/ComfyUI_Custom_Nodes_AlekPet

  • - GoogleTranslateTextNode (2)

  • - CM_FloatBinaryOperation (2)

  • - CM_FloatToInt (2)

  • - CM_IntToFloat (2)

  • - CM_SDXLResolution (1)

ComfyUI

  • - PreviewImage (3)

  • - UNETLoader (1)

  • - DualCLIPLoader (1)

  • - VAELoader (2)

  • - EmptyLatentImage (1)

  • - LoraLoader (3)

  • - KSamplerSelect (1)

  • - LatentUpscaleBy (1)

  • - BasicGuider (1)

  • - RandomNoise (1)

  • - VAEDecode (2)

  • - SamplerCustomAdvanced (1)

  • - BasicScheduler (1)

  • - SaveImage (3)

  • - ImageScale (1)

  • - LoadImage (1)

  • - MiDaS-DepthMapPreprocessor (1)

  • - HEDPreprocessor (1)

  • - CannyEdgePreprocessor (1)

  • - AutoNegativePrompt (1)

  • - UltimateSDUpscaleNoUpscale (1)

  • - UltimateSDUpscale (1)

  • - Upscale Model Loader (1)

Checkpoints (0)

LoRAs (10)

FLUX_BLACK MYTH WUKONG黑神话:悟空.safetensors

FLUX_建筑风景V1.safetensors

FLUX超写实逼真黑悟空.safetensors

anime_lora_comfy_converted.safetensors

art_lora_comfy_converted.safetensors

disney_lora_comfy_converted.safetensors

furry_lora.safetensors

mjv6_lora_comfy_converted.safetensors

realism_lora_comfy_converted.safetensors

scenery_lora_comfy_converted.safetensors