The Little Mermaid Cartoon Blind Box


Updated:

Workflow Preview

Showcases

Workflow name:

Little Mermaid Cartoon Blind Box

[Workflow Introduction]

cloud running url:

The Little Mermaid Cartoon Blind Box

This workflow requires uploading two images, a mermaid-style image and a portrait. Then, by using the Joy_caption node, you can provide reverse prompt words for the image, and finally give it to the sampler through the model to get the final image.

[Usage scenario]

You can generate an exclusive mermaid-style blind box through two images. Unleash your creativity and create your own exclusive blind box.

[Key nodes]

FLUX model

[Model version]

FLUX Model name: flux1-dev-fp8-Comfy-Org.safetensors

[LoRA model]

Model name: maghe_rank16_bf16.safetensors

[ControlNet application]

Model name: flux-canny-controlnet-v3-xlabs-dev.safetensors

[K sampler]

CFG: 1.0

Sampling method: euler

Scheduler: simple

Noise reduction: 1.00

Nodes Detail

44 Nodes
Primitive Node Types
19
  • CLIPTextEncode1
  • VAEEncode1
  • KSampler2
  • VAEDecode2
  • SaveImage1
  • PreviewImage5
  • ControlNetLoader2
  • CheckpointLoaderSimple1
  • LoadImage2
  • LoraLoader1
  • EmptyLatentImage1
Custom Node Types
25
  • PulidFluxEvaClipLoader1
  • PulidFluxInsightFaceLoader1
  • PulidFluxModelLoader1
  • LayerUtility: TextJoin1
  • Joy_caption_load1
  • FluxGuidance1
  • ConditioningZeroOut1
  • ApplyPulidFlux1
  • easy imageSize1
  • EmptySD3LatentImage1
  • DownloadAndLoadDepthAnythingV2Model1
  • DepthAnything_V21
  • ImageScaleToTotalPixels1
  • AIO_Preprocessor1
  • ImageResizeKJ1
  • CR Text1
  • easy showAnything1
  • Joy_caption1
  • Crop Face1
  • Load RetinaFace1
  • easy cleanGpuUsed1
  • Label (rgthree)2
  • ControlNetApplySD32
Comments