The purpose of this workflow is to mix flux2klein excellent prompt tracking and high image quality with illustrious- SDXL's vast AGC character resources.
Runtime profile
Source description
flux2_klein-SDXL_Upscale_workflow
(text to image upscale)
The purpose of this workflow is to mix flux2_klein excellent prompt tracking and high image quality with illustrious- SDXL's vast AGC character resources.
In other words, characters that flux2_klein has not yet learned can be expressed using illustrious-SDXL's learned resources without using Lora.
It combines flux2_klein and SDXL for extremely fast generation speeds .
This is how I use it, but it's up to you to use it in your own way.
AI-generated commentary
AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.
This workflow is for text-to-image generation with optional upscaling: it combines flux2_klein prompt tracking with illustrious-SDXL character resources, takes a text prompt, and produces a rendered image or an upscaled image.
For characters that flux2_klein has not learned, it can use illustrious-SDXL's learned resources without a LoRA.
You can also use it as a regular flux2_klein workflow.
Provide a text prompt to generate a rendered image; the workflow also supports an upscaled image.
Use ComfyUI version 0.9.1 or higher.
The named node packs are ComfyUI-Custom-Scripts (1.2.5), ComfyUI-Easy-Use (9947b8be70e17ceabafa4fa7f476b54626de4704), comfyui-mute-bypass-by-ID (2.0.0), and images-grid-comfy-plugin (2.6.0).
The named model files include flux-2-klein-4b-fp8.safetensors, qwen_3_4b.safetensors, flux2-vae.safetensors, and dmd2_sdxl_4step_lora_fp16.safetensors.
It also names lollypop.pth and Luc_meqemeqe-SR.safetensors.
Turn off Upscale-Switch to render only flux2_klein; in that mode, the workflow does not upscale anything.
For upscaling, use a ratio from 1.5 to 2.0; 1.5 is recommended.
Suggestion · not verified
Verify whether your setup should use flux-2-klein-4b-fp8.safetensors or flux-2-klein-4b.safetensors.
Suggestion · not verified
Before use, verify the intended files for the text encoder, VAE, upscaler, and checkpoint: qwen_3_4b.safetensors, flux2-vae.safetensors, lollypop.pth, and Luc_meqemeqe-SR.safetensors.
Suggestion · not verified
Does this need to be edited?
Sign in to send an edit request.
Sources
1 sourceSource excerpts
1 excerptSource context: 354 downloads · Type Workflows · Base model Flux.2 D
Turn Off Upscale-Switch, only flux2_klein will be rendered.
Can be used as a regular flux2_klein workflow
Doesn't upscale anything
Upscale ratio 1.5 to 2.0 (1.5 recommended)
Requirements
ComfyUI ver. 0.9.1 or higher
flux2_klein
Models
flux-2-klein-4b-fp8.safetensors
or
flux-2-klein-4b.safetensors
text_encoders
VAE
Illustrious Model
Civitai | Share your models Illustrious Checkpoint
also recommend my merge model
DMD2 Speed LoRA
dmd2_sdxl_4step_lora_fp16 or dmd2_sdxl_4step_lora_fp32
Recommend upscaler lollypop
Use Custom Node
Estimated VRAM requirement
Estimate unavailable
4.16 GB across 2 of 6 model files. Model file total + 25% loading overhead + 2 GB execution buffer, rounded up.
Requirements
10 requirementsLoRA · 376 MB · SAFETENSOR · Unknown
Checkpoint · 3.79 GB · SAFETENSORS · fp8 · Unknown
flux2-vae.safetensors
PossibleVAE · Unknown
lollypop.pth
Not resolvedUpscaler · Unknown
Luc_meqemeqe-SR.safetensors
Not resolvedCheckpoint · Unknown
Node pack · Registry
Node pack · Registry
Node pack · Registry
Node pack · Registry