FLUX.2 KLEIN9B — Custom Merge
Model Overview
A custom merged model based on FLUX.2 [klein] 9B by Black Forest Labs. This model has been tested and optimized for high-quality realistic image generation with exceptional detail in textures, lighting, and atmospheric effects.
Key Features
Photorealistic skin rendering with natural subsurface scattering, visible pores, and subtle imperfections
Exceptional water and wet surface handling — droplets, reflections, refraction, and flow patterns
Advanced lighting capabilities including volumetric god rays, golden hour, rim lighting, and chiaroscuro
Superior hair detail with individual strand rendering and natural backlighting effects
Rich material textures — velvet, linen, silk, leather, wood, metal, and stone
Strong atmospheric effects — fog, mist, steam, rain, and environmental depth
Anatomical accuracy for both portrait and full-body generation
Recommended Settings
Optimal Configuration (Tested):
CFG Scale: 2.0
Steps: 15
Scheduler: Euler Ancestral (EulerA)
Quick Previews:
CFG Scale: 1.5
Steps: 8-10
Scheduler: EulerA
Complex Scenes (Landscapes, Materials):
CFG Scale: 2.0-2.5
Steps: 15-20
Scheduler: EulerA
Resolution: 1024x1536 (portrait) or 1536x1024 (landscape)
Best Use Cases
Photorealistic portraits and character art
Environmental and landscape photography
Fashion and material studies
Cinematic and atmospheric scenes
Fine art nude photography (artistic interpretation)
Macro detail shots
Water and weather effects
Known Limitations
Transparent/translucent fabrics may require LoRA or enhanced prompting for proper rendering
Text rendering is readable but not perfect — acceptable for most use cases
Complex multi-element scenes may require 2-3 generations for optimal results
Official 4-step inference is too low for detailed work — 15 steps recommended
Technical Details
Base Model: FLUX.2 [klein] 9B by Black Forest Labs
Architecture: Rectified Flow Transformer
Type: Step-distilled model (optimized for speed)
License: Refer to original FLUX.2 license terms
Performance Notes
This model excels at organic, natural scenes with complex lighting. The EulerA scheduler at 15 steps provides the best balance between speed and quality. While the official recommendation is 4 steps at CFG 1.0, extensive testing shows 15 steps at CFG 2.0 delivers significantly better results for detailed work without sacrificing the speed advantage of distilled models.
Credits
Base model by Black Forest Labs. Custom merge and testing by community contributors.
Enjoy creating! If you generate something amazing, consider sharing it in the images section. 🎨
Description
more add ones
Tested ONLY ForgeNeo
Otger UI at your own risk
FAQ
Comments (14)
Not working in ComfyUI
In most cases, the breakdown is due to insufficient GPU memory. The FP16 model and encoder are very heavy.
It‘s not working in my Comfy too, and I have a lot of VRAM and RAM. ;)
Try it as lora.
@LonelyCoyote Ok, and what should I use as Diff? It can’t be empty…
yes, this unfortunately has never worked in comfyUI for me. All updated, 5090, 128GB memory, 4TB SSD. It's not a resource issue.
@LetTheBassDrop Yeah, same specs on my side. It‘s DEFINITELY not a resource problem. Something has to be really wrong with the layers, I guess.
@CupRunethOverWithLSD Flux-2-klein-9b
@CupRunethOverWithLSD This is the part of the file header:
{"format": "pt",
"ss_network_module": "lora",
"ss_base_model_version": "flux2_klein_9b",
"modelspec.title": "Extracted LoRA - Unchained-V2-64",
"software": "ai-toolkit v0.7.23"
}
It is LoRA.
@LonelyCoyote Wait a moment - the model only uses Unchained-Lora? Nothing more? XD xD
@LonelyCoyote there is no Lora „Unchained V2“…
@CupRunethOverWithLSD I don't know how it was prepared or what the author used. I simply wrote what I see in the header.
@LonelyCoyote try this take, its checpoint done with another merging process https://civitai.com/models/2387901?modelVersionId=2776701
Oh, sorry for the delay, I just now found time to check the web server. Your latest version is working fine in Comfoy. Thank you so much for this wonderful model!