Lock in a year of flow. Get 50% off your first year. Limited time offer. Claim now ⏰

ThinkDiffusion

Workflows

Product

Pricing

Enterprise

Docs

Lock in a year of flow. Get 50% off your first year. Limited time offer. Claim now ⏰

ThinkDiffusion

Lock in a year of flow. Get 50% off your first year. Limited time offer. Claim now ⏰

Text to Character Sheet with a reference LoRA

Generate a character sheet using a prompt and a LoRA model of a particular person for more accurate renders. Key Inputs Load Image: Use any JPG or PNG of your pose sheet Prompt: as descriptive a prompt as possible Width & height: Optimal resolution settings are noted at 1280px x 1280px Denoise: The amount of variance in the new image. Higher has more variance. ControlNet Strength: The amount of adherence to the original image. Higher has more adherence. Start Percent: The point in the generation process where the control starts exerting influence. (Have it start later, to let AI imagine first.) End Percent: The point in the generation process where the control stops exerting influence. (Have it end sooner, to let AI finish it off with some variation.) Flux Guidance: How much influence the prompt has over the image. Higher has more guidance.

Character Sheet

Controlnet

Flux

229

Generate a character sheet using a prompt and a LoRA model of a particular person for more accurate renders.

Key Inputs

Load Image: Use any JPG or PNG of your pose sheet

Prompt: as descriptive a prompt as possible

Width & height: Optimal resolution settings are noted at 1280px x 1280px

Denoise: The amount of variance in the new image. Higher has more variance.

ControlNet Strength: The amount of adherence to the original image. Higher has more adherence.

Start Percent: The point in the generation process where the control starts exerting influence. (Have it start later, to let AI imagine first.)

End Percent: The point in the generation process where the control stops exerting influence. (Have it end sooner, to let AI finish it off with some variation.)

Flux Guidance: How much influence the prompt has over the image. Higher has more guidance.

Uploading LoRAs to Floyo

Launch App or Run Workflow
On the left side of the canvas, click the middle file button
In "My Files" navigate to models/loras, select "Upload a File" and choose a LoRA file
Once your LoRA is uploaded, close the file browser and refresh your ComfyUI canvas by pressing "R", this will update all model lists
Select your LoRA in the Load LoRA node

EXTENSIONS

rgthree-comfy

rgthree

rgthree-comfy enhances COMFYUI by introducing various nodes and improvements that streamline workflows. Key features include intuitive seed control, advanced rerouting options, and efficient node management tools, boosting user experience and output quality.

2506

2025-12-08

ComfyUI Impact Subpack

ltdrdata

This extension enhances the ComfyUI Impact Pack by introducing the UltralyticsDetectorProvider, which supports segmentation and bounding box detection models, improving detection capabilities within the COMFYUI framework.

224

2025-07-14

ComfyUI-WJNodes

807502278

ComfyUI-WJNodes is a versatile node package for COMFYUI, offering various functionalities such as image loading, editing, cropping, and batch processing, enhancing user workflows and output quality.

2025-07-02

UltimateSDUpscale

ssitu

ComfyUI_UltimateSDUpscale adds specialized nodes for enhancing image quality within COMFYUI, leveraging Coyote-A's upscale script. Key features include various upscaling options and customizable sampling parameters for improved image processing.

1237

2025-07-16

ComfyUI-Impact-Pack

ltdrdata

COMFYUI's Impact Pack enhances image processing with custom nodes for detection, detail refinement, and upscaling. It integrates seamlessly, offering features like segmentation detection and advanced detailers for improved image quality.

2770

2025-11-03

ComfyUI Official

comfyanonymous

ComfyUI is a versatile graphical interface for building and managing complex Stable Diffusion workflows, utilizing a node-based system for enhanced user experience and flexibility.

87939

2025-09-09

Generates in about 5 mins 37 secs

floyoofficial

Nodes & Models

Generate a character sheet using a prompt and a LoRA model of a particular person for more accurate renders.

Key Inputs

Load Image: Use any JPG or PNG of your pose sheet

Prompt: as descriptive a prompt as possible

Width & height: Optimal resolution settings are noted at 1280px x 1280px

Denoise: The amount of variance in the new image. Higher has more variance.

ControlNet Strength: The amount of adherence to the original image. Higher has more adherence.

Start Percent: The point in the generation process where the control starts exerting influence. (Have it start later, to let AI imagine first.)

End Percent: The point in the generation process where the control stops exerting influence. (Have it end sooner, to let AI finish it off with some variation.)

Flux Guidance: How much influence the prompt has over the image. Higher has more guidance.

Uploading LoRAs to Floyo

Launch App or Run Workflow
On the left side of the canvas, click the middle file button
In "My Files" navigate to models/loras, select "Upload a File" and choose a LoRA file
Once your LoRA is uploaded, close the file browser and refresh your ComfyUI canvas by pressing "R", this will update all model lists
Select your LoRA in the Load LoRA node