Civitai #145846611: Night Onsen Rotenburo Portrait - RedCraft Hybrid H3 & Krea2T Photorealism
Full Resolution
Generation Specs RTX 4070
Sampler euler
Scheduler simple
Steps 8
CFG Scale 1
Seed 2356125508
Target VRAM 12GB
VRAM 10GB - 12GB
GPU Compatibility
Optimal Performance

Runs entirely in VRAM with no memory swapping.

RedCraft | 红潮 | Hybrid H3 & Krea2T & QW21 Civitai #145846611 Min 10GB VRAM 896×1600 RTX 4070 Verified

Night Onsen Rotenburo Portrait - RedCraft Hybrid H3 & Krea2T Photorealism

Atmospheric nighttime outdoor hot spring lookbook portrait deconstructed from Civitai #145846611 with White-Weapon SFW aesthetics and RedCraft Hybrid H3 & Krea2T.

Blueprint Summary RTX 4070 Verified

Reproducible ComfyUI workflow for Civitai #145846611: Night Onsen Rotenburo Portrait - RedCraft Hybrid H3 & Krea2T Photorealism using RedCraft | 红潮 | Hybrid H3 & Krea2T & QW21 at 896×1600 resolution. Requires minimum 10GB VRAM with sampler euler and scheduler simple (8 steps). Includes 1-click terminal model sync and canvas JSON graph.

Pipeline Flow 9 Nodes
Open in Resolver
01 Load Models
UNETLoader
3 loaders (DiT, CLIP, VAE)
02 Conditioning
CLIPTextEncode
2 prompt encodings
03 KSampler
KSampler
DiT latent denoising
04 Decode & Save
VAEDecode
Latent to pixel space

Execution DAG Topology Interactive Visualizer

Drag to pan · Scroll to zoom · Hover wires
Node Graph 0 nodes
MODEL CLIP LATENT VAE IMAGE

Model & Asset Setup Setup Script

Run in your ComfyUI root:

curl -fsSL https://decomfy.com/api/scripts/civitai-onsen-night-rotenburo-redcraft.sh | bash

Positive Prompt

Full-body vertical 9:16 portrait of a cheerful young Japanese woman (22) with glossy dark hair and natural bangs, playfully crouching low on the flat river stones beside a steaming outdoor Japanese onsen (rotenburo) at night. She has a bright, radiant smile, facing the camera with both hands raised and fingers spread wide in an energetic, joyful candid gesture. She is wearing a chic minimalist black textured one-piece resort swimwear with sleek halter straps, paired with stylish black lace-up leather ankle boots with slender heels. In the background, rising mist and steam curl off the dark reflective thermal hot spring pool, bordered by weathered natural rocks and a traditional cedar bamboo fence, softly illuminated by the warm amber glow of a stone paper lantern (andon). 35mm film photography, crisp focal clarity on the subject, authentic skin texture with natural pores and gentle specular highlights from the humid atmosphere, atmospheric nighttime hot spring ambiance.

Negative Prompt

unclothed, disrobed, revealing clothes, lowres, blurry, bad anatomy, deformed hands, extra fingers, low quality, artifacts, distorted proportions, text, watermark, signature
Prompt Customization: This workflow is pre-wired and calibrated. Swap this prompt with your own character, subject, or style without breaking the pipeline.

Required Models 3 Models

Required Storage: 17.7 GB (3 models · DiT/Base: 12.6 GB · Text Encoder: 4.9 GB · VAE: 253 MB)
models/diffusion_models/ 12.56 GB Civitai
redcraftHybridH3Krea2t_40K2TDUAL_3196558.safetensors Official RedCraft | Hybrid H3 & Krea2T DUAL MASO W6A8 (Model #958009, Version #3311198, 0 LoRAs Required)
models/text_encoders/ 4.88 GB HuggingFace
qwen3vl_4b_fp8_scaled.safetensors Official Qwen3-VL 4B FP8 Scaled Text Encoder (CLIPLoader type: krea2)
models/vae/ 253 MB HuggingFace
qwen_image_vae.safetensors Official Qwen Image VAE

Field Notes RTX 4070 Benchmark

Benchmark: Rendered on NVIDIA GeForce RTX 4070 (12GB VRAM, CUDA 12.4): 8 steps Euler Simple @ CFG 1.0, 896x1600 vertical resolution, peak 11.75 GB VRAM
1. CIVITAI PROVENANCE & HYBRID ARCHITECTURE ANALYSIS:\nForensic extraction of Civitai image #145846611 by creator vorpal_milf revealed an advanced model recipe: RedCraft 3.0 (FixQW21T) combined with FineView QW21 Turbo Fixed (Model #3003597, Version #3407540). The architecture fuses Krea 2 Turbo latent dynamics with Qwen 2.1 DiT transformer blocks to achieve exceptional photorealistic human proportions, intricate finger articulation, and natural facial expressions at high vertical resolutions (896x1600).\n\n2. WHITE-WEAPON SFW RE-ENGINEERING & UNIVERSAL CLEAN NEGATIVES:\nThe original creator artwork contained unconstrained adult nudity. Applying the White-Weapon Protocol ("T\u1ea9y Tr\u1eafng"), we anchored positive clothing structures in a chic minimalist black textured one-piece resort swimwear with clean architectural halter straps, preserving the exact subject silhouette, crouching pose, hand gestures, and stylish black lace-up leather ankle boots. The steaming outdoor rotenburo pool, weathered river rocks, rustic cedar bamboo enclosure, and warm amber illumination from the paper andon lantern were rendered with 100% atmospheric fidelity. Negative conditioning strictly utilizes Universal Clean Negatives, completely free of raw adult tokens to safeguard Google SafeSearch indexing.\n\n3. HARDWARE BENCHMARKS & SIGNAL FLOW ON RTX 4070:\nExecuted natively on external GPU host wingpu (NVIDIA GeForce RTX 4070 12GB VRAM, CUDA 12.4). The pipeline feeds UNETLoader with quantized weights (redcraftHybridH3Krea2t_40K2TDUAL_3196558.safetensors, fp8_e4m3fn) directly to Qwen3-VL 4B text encoder and Qwen Image VAE. Latent sampling runs via Euler Simple across 8 steps at CFG 1.0, consuming 11.75 GB peak VRAM. Zero external LoRAs are required, ensuring zero missing node errors and instantaneous 1-click execution for developers.

Frequently Asked Questions FAQ

What GPU and VRAM are required to run Civitai #145846611: Night Onsen Rotenburo Portrait - RedCraft Hybrid H3 & Krea2T Photorealism?

This workflow requires a minimum of 10GB VRAM (recommended 12GB VRAM). Tested and verified on NVIDIA GeForce RTX 4070 (12GB VRAM) at 896x1600 resolution.

How do I resolve missing custom nodes for this workflow?

You can drop the workflow JSON into our client-side Missing Node Auto-Resolver at https://decomfy.com/resolve/ to detect missing nodes and generate install commands, or run the 1-click terminal setup script provided below.

Which VAE and conditioning text encoders are required for Krea 2 DiT?

Krea 2 DiT single-stream models strictly require the 16-channel Wan2.1 VAE and Qwen3-VL text encoder. Using standard Flux or SDXL VAE causes inverted checkerboard artifacts. Also pair with RescaleCFG for high-contrast portrait clarity.

Related ComfyUI Blueprints

View all 52 workflows
Krea 2 DiT
DiT Blueprint #145657534

Sick Ollie & Marla - Nostalgic Retro Lookbook (Krea 2)

9st · euler 1080×1440 10GB
Krea 2 DiT
DiT Blueprint #145646258

RedCraft 4.0 - Intimate Editorial Lookbook (K2T DUAL)

8st · euler 1888×3328 10GB
Krea 2 DiT
Fast Turbo / DMD2 #145349049

High-Contrast Penthouse Skyline Silhouette - Krea 2 Turbo & Qwen3-VL

9st · er_sde 944×1664 10GB
Action completed