1. THE LATEST
Hot off the model press from Blackforest Labs!
FLUX.2 Klein 9B for Text to Image
Create a high quality image using 9B model of Flux 2 Klein
FLUX.2 Klein 9B for Image Editing
Unified workflow: one model for text‑to‑image, image‑to‑image, and image editing
FLUX.2 Klein 9B Image Inpainting
Inpainting image using Flux.2 Klein and LanPaint
2. CORE FUNCTIONALITY
Workflows based on core capabilities.
Flux 2 Text-to-Image Generation
Flux Kontext Multi-Image Reference
Combine up to 3 reference images into one with Flux Kontext Key Inputs Load Image (3x): Load 3 different reference images. Prompt: Describe how to combine these images, see default value for example.
Flux Dev - Text to Image w/ Optional Image Input
Flux Text to Image
Create original images using only text prompts, which can be simple or elaborate. Key Inputs Prompt: as descriptive a prompt as possible Width & height: Optimal resolution settings are noted
Flux Kontext Inpainting
Inpaint with the power of Flux Kontext, providing additional content-awareness.
Wan2.1 Fun Control and Flux for V2V Restyle
Create a new video by restyling an existing video with a reference image.
Flux Kontext - Quick & Easy
Load an image reference and use the smart Flux Kontext model to ask for anything. The model understands natural language and looks at your input image. Example: Put this man on a tropical island. This man is sleeping in a bed.
SMART FACE SWAPPER - Ace++ Flux Face swap
Face swapper built with Flux and ACE++. Works with added details too, like hats, jewelry. Smart features, use natural language.
Flux Inpaint - ULTIMATE workflow.
Flux inpainting with Flux fill model and regular Flux dev model. All in one Flux inpainting workflow that easily AI retouches any image.
Flux Text to Character Sheet
Create a character and a range of consistent outputs suitable for establishing character consistency, training a model, and ensuring consistency throughout multiple scenes. Key Inputs Image reference: Use the included pose sheet to show range of positions Prompt: as descriptive a prompt as possible
Flux Fill Dev Image Outpainting
Extend your images out for a wider field of view or just to see more of your subject. Expand compositions, change aspect ratios, or add creative elements while maintaining consistency in style, lighting, and detail while seamlessly blending with the existing artwork. Enhance visuals, create immersive scenes, and repurpose images for different formats without losing their original essence. Key Inputs Image reference: Use any JPG or PNG showing your subject clearly Prompt: as descriptive a prompt as possible to describe the area you want to extend out to Left, Right, Top, Bottom: Amount of extension in pixels Feathering: Amount of radius around the original image in pixels that the AI generated outpainting will blend with the original
Image Redux with Flux
Create variations of a given image, or restyle them. It can be used to refine, explore, or transform ideas and concepts. Key Inputs Image reference: Use any JPG or PNG showing your subject clearly Width & height: In pixels Prompt: as descriptive a prompt as possible Strength (step 5: value): Strength of redux model, play around with the value to increase or decrease the amount of variation
Flux Simple Inpainting
3: Customizations.
Workflows that combine other models, and extend Flux capabilities beyond core functionality.
Flux ControlNet 2.0 - All-in-one
Latest and greatest Flux ControlNet workflow.
Image to Image with Flux ControlNet
Transform your images into something completely new, yet retaining specific details and composition from your original using flexible controls. Key Inputs Image reference: Use any JPG or PNG showing your subject clearly Prompt: as descriptive a prompt as possible Denoise Strength: The amount of variance in the new image. Higher has more variance. Width & height: Try and match the aspect ratio of the original if possible.
Flux Text2Image w/ 2 Tone Color Grade
Generate cinematic images using Flux with the simple workflow to choose color tint in highlights and shadows Key Inputs Prompt: Set text prompt SDXLResolution: Choose your aspect ratio/resolution Seed: Choose either a fixed or random seed (default value: fixed) Highlights/Shadows: Choose color grade
Qwen Image Edit 2509 + Flux Krea for Creating Next Scene
Flux Kontext Sketch to LineArt + Color Previz
Quickly convert rough sketches into polished lineart and colorized concepts. Ideal for early storyboards, character designs, scene planning, and other visual explorations.
Flux Kontext - Sketch to Image
Bring your sketches to life in full color with Flux Kontext! Key Inputs Load Image – Upload the sketch you want to transform. Prompt – Describe the desired output style, such as: “Render this sketch as a realistic photo” or “Turn this sketch into a watercolor painting.”
Multi-Image Flux Ultra, Pro, Dev, Recraft+
Start with a prompt, and get a different render from a range of unique models at the same time.
Flux Outfit Transfer
Virtual Outfit Try-On with Auto Segmentation Try virtual clothing on any subject using Flux Dev, Ace Plus, and Redux, with automatic segmentation. Great for concept previews, fashion mockups, or character styling. Key Inputs Outfit: Load the outfit image you want to apply. Make sure it's high quality — visible artifacts or distortions may carry over into the final result. Actor: Add the subject or character you want to dress. Ideally, use a clear, front-facing image. Human Parts Ultra: Choose which parts of the body the clothing should apply to. For example, for a long-sleeve shirt, select: torso, left arm, and right arm. This helps the model align the clothing properly during generation. Prompt: Default value works for most outfits, however you may try to adjust it to describe the desired outfit.
_1758863167371.webp?width=400&height=300&quality=80&resize=cover)
Flux Kontext and HD360 LoRA for 360 Degree View
Flux Kontext 360° Workflow - Seamless Panorama Generation Input: Simply upload an image in the "Load Image from Outputs" node Output: A 360° Panoramic image
Flux Image Upscaler with UltimateSD
A simple workflow to enlarge & add detail to an existing image. Key Inputs Image: Use any JPG or PNG Upscale by: The factor of magnification Denoise: The amount of variance in the new image. Higher has more variance.
4. LoRA TRAINING & USAGE.
Workflows to train and use Flux.
Fast LoRA Training for Flux via Floyo API
FLUX is great at generating images, but locking in a specific aesthetic or character is easier with a LoRA. Here's how to create your own.
ComfyUI Flux LoRA Trainer
Created by @Kijai on Github, please support the original creator!
Flux Character LoRA Test and Compare
Test and compare multiple epochs of a character LoRA side by side with preset prompts When training a LoRA, you'll usually have a few checkpoints throughout the process to test. This workflow lets you load up to 4 LoRAs to test side by side, making it easier to determine which one is right for you! Key Inputs: LoRA Loaders: Load each LoRA epoch for the same character in up to 4 groups. Groups Bypasser: Enable/disable groups as needed. If you only have 2 epochs to test, disable the back 2 groups! Triggerword: Simply add the trigger word for your LoRA and it will auto-fill in the default prompts. Leave blank if you're using your own custom prompts that include the trigger word. LoRA Testing Prompts: Default prompts work well to get an idea of how your character will look in different situations, but feel free to replace them with your own prompts (max 4).
Flux Kontext - Single Image to Character LoRA
Generate a 60-Image LoRA Dataset from a Single Character Image
Text to Image + LoRA model
Create an image from a trained AI model of something specific ( a specific figure, outfit, art style, product etc) to ensure specific details within.
Text to Image with Multi-LoRA
Create consistent images with multiple LoRA models.
Image Inpainting with LoRA
Change specific details on just a portion of the image for inpainting or Erase & Replace, adding a LoRA for extra control.
Image Upscaler with LoRA
Create a larger more detailed image along with an extra AI model for fine tuned guidance. Key Inputs Load Image: Use any JPG or PNG showing your subject clearly Load LoRA: Load your reference model here Prompt: as descriptive a prompt as possible Upscale by: The factor of magnification Denoise: The amount of variance in the new image. Higher has more variance.