Skip to content
WorkflowListedExternal
Open source

Flux Kontext workflow - detailers, IPAdapter, upscale, all optional — v1.0

Nodes you will probably want to adjust and use normally.

Sourcecivitai.com/models/1766966
TaskMultiCapabilitiesImage to image · Text to image · UpscalingFamilyFLUXEstimated VRAMNot yet characterizedBgrade

Runtime profile

Access
External workflow
Status
Listed
Source
Civitai
Updated
Sep 1, 2026

Source description

About the workflow:

Nodes you will probably want to adjust and use normally.

Init Load the pictures to be used with Kontext.

Loader Select the diffusion model to be used, as well as load CLIP, VAE and select latent size for the generation.

Prompt Pretty straight forward: your prompt goes here.

Switches Basically the "configure" group. You can enable / disable model sampling, LoRAs, detailers, upscaling, automatic prompt tagging, clip vision UNClip conditioning and IPAdapter. I'm not sure how well those last two work, but you can play around with them.

Model settings Model sampling and loading LoRAs.

Sampler settings Adjust noise seed, sampler, scheduler and steps here.

1st pass The generation process itself with no upscaling.

Upscale The upscaled generation. By default it makes a factor of 2 upscale, with 2x2 tiled upscaling.

Mess with these nodes if you like experimenting, testing things.

Conditioning Worthy to mention that FluxGuidance node is located here.

Detail sigma Detailer nodes, I can't easily explain what does what, but if you're interested, look the nodes' documentation up. I set them at a value that normally generates the best results for me.

Clip vision and IPAdapter Worthy to mention that I have yet to test how well ClipVision works and IPAdapter's strength when it comes to Flux Kontext.

What you need

Model files
9.7 GiB known subtotal

Remaining hardware or download-size estimates are not available yet.

Requirements

Models

10

  • 4x-UltraSharp.pth

    Found

    Upscaler · 63.9 MB · PTH · Unknown

  • ae.safetensors

    Possible

    VAE · Unknown

  • clip_l.safetensors

    Possible

    Text encoder · Unknown

  • clip_vision_h.safetensors

    Possible

    Vision model · Unknown

  • comfyui_ultimatesdupscale

    Not resolved

    Upscaler · Registry

  • Dark_Ghibli.safetensors

    Possible

    LoRA · Unknown

  • flux1-kontext-dev.safetensors

    Found

    Checkpoint · 22.2 GB · SAFETENSORS · fp16 · full · Unknown

  • kontext-turnaround-sheet-v1.safetensors

    Not resolved

    LoRA · Unknown

  • svdq-int4_r32-flux.1-kontext-dev.safetensors

    Found

    Checkpoint · 6.30 GB · SAFETENSOR · fp16 · Unknown

  • t5xxl_fp16.safetensors

    Possible

    Text encoder · Unknown

Node packs

6

AI-generated commentary

AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.

Purpose

  • This Flux.1 Kontext workflow generates images from loaded pictures and a text prompt, using a selected diffusion model, CLIP, VAE, latent size, and sampler settings; it produces a first-pass generation and can produce an upscaled generation, with switches for detailers, LoRAs,-

Flow

  • Load pictures in Init, select the diffusion model, CLIP, VAE, and latent size in Loader, then enter your prompt.

  • The first pass generates without upscaling; the Upscale stage uses a factor of 2 with 2x2 tiled upscaling by default.

Inputs and outputs

  • Inputs include loaded pictures, a text prompt, the diffusion model, CLIP, VAE, latent size, noise seed, sampler, scheduler, and steps.

  • Outputs include a first-pass generation and, when upscaling is enabled, an upscaled generation.

Setup

  • For detailer experiments, consult each node's documentation because the workflow description does not explain what each detailer node does.

    Suggestion · not verified

Dependencies

  • Node packs listed for this workflow are comfy-image-saver, comfyui_ultimatesdupscale, ComfyUI-Crystools, comfyui-detail-daemon, comfyui-easy-use, ComfyUI-nunchaku, and comfyui-wd14-tagger.

  • Model files listed for this workflow are 4x-UltraSharp.pth, ae.safetensors, clip_l.safetensors, clip_vision_h.safetensors, Dark_Ghibli.safetensors, flux1-kontext-dev.safetensors, kontext-turnaround-sheet-v1.safetensors, svdq-int4_r32-flux.1-kontext-dev.safetensors, and t5xxl_fp16

Configuration

  • Switches can enable or disable model sampling, LoRAs, detailers, upscaling, automatic prompt tagging, CLIP Vision/unCLIP conditioning, and IPAdapter.

  • Model settings cover model sampling and LoRA loading; Sampler settings expose noise seed, sampler, scheduler, and steps, while Conditioning includes FluxGuidance.

Cautions

  • Before using it, verify that all listed model files and node packs are available.

    Suggestion · not verified

  • Also verify that Dark_Ghibli.safetensors and kontext-turnaround-sheet-v1.safetensors are the intended files before using them.

    Suggestion · not verified

Does this need to be edited?

Sign in to send an edit request.

Sources

1 source
CivitaiPrimary
civitai.com/models/1766966

Verified Jul 19

Source excerpts

1 excerpt
NoteJul 19civitai.com/models/1766966

Source context: 4802 downloads · Type Workflows · Base model Flux.1 Kontext