Civitai #145657534: Sick Ollie & Marla - Nostalgic Retro Lookbook (Krea 2)

Full Resolution

Generation Specs RTX 4070

Sampler euler

Scheduler beta

Steps 9

CFG Scale 1

Seed 708863480517334

Target VRAM 12GB

VRAM 10GB - 12GB

GPU Compatibility

Optimal Performance

Runs entirely in VRAM with no memory swapping.

Sick Ollie (Krea 2 FP8) Civitai #145657534 Min 10GB VRAM 1080×1440 RTX 4070 Verified

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

Photorealistic retro-bedroom editorial lookbook reproduced natively with Sick Ollie Krea 2 FP8 and Marla LoRA adapter.

Blueprint Summary RTX 4070 Verified

Reproducible ComfyUI workflow for Civitai #145657534: Sick Ollie & Marla - Nostalgic Retro Lookbook (Krea 2) using Sick Ollie (Krea 2 FP8) at 1080×1440 resolution. Requires minimum 10GB VRAM with sampler euler and scheduler beta (9 steps). Includes 1-click terminal model sync and canvas JSON graph.

Pipeline Flow 10 Nodes

Open in Resolver

01 Load Models

UNETLoader

4 loaders (DiT, CLIP, VAE)

02 Adapters

LoraLoader

1 adapter

03 Conditioning

CLIPTextEncode

2 prompt encodings

04 KSampler

KSampler

DiT latent denoising

05 Decode & Save

VAEDecode

Latent to pixel space

Execution DAG Topology Interactive Visualizer

Drag to pan · Scroll to zoom · Hover wires

Model & Asset Setup Setup Script

Run in your ComfyUI root:

curl -fsSL https://decomfy.com/api/scripts/civitai-marla-retro-lookbook-krea2.sh | bash

Positive Prompt

dr0smarla, high-fashion editorial portrait of a beautiful young woman standing gracefully beside a vintage wooden writing desk by a sunlit window, wearing an elegant tailored floral linen sundress. Venetian-blind shadows create soft cinematic light stripes across the warm room, a delicate vase of fresh wildflowers and vintage notebooks on the desk nearby. She glances gently over her shoulder with a relaxed, confident expression. Warm nostalgic interior, 35mm film aesthetic, shallow depth of field, natural skin texture, cozy retro-bedroom daylight lookbook.

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.

LoRA Adapter Stack 1 Adapters

RTX 4070 Calibrated Weights

01 dr0smarla_krea2_v1.safetensors +1

Required Models 4 Models

Required Storage: 17.9 GB (4 models · DiT/Base: 12.5 GB · Text Encoder: 4.9 GB · LoRAs: 223 MB · VAE: 253 MB)

models/diffusion_models/ 12.52 GB Civitai

sickOllie_krea2_3035266.safetensors Official Sick Ollie Krea 2 FP8 Checkpoint (Model #2676616, Version #3154408)

Get

models/loras/ 223 MB Civitai

dr0smarla_krea2_v1.safetensors Marla Character LoRA for Krea 2 (Trigger: dr0smarla, Model #2997563, Version #3399901)

Get

models/text_encoders/ 4.88 GB HuggingFace

qwen3vl_4b_fp8_scaled.safetensors Official Qwen3-VL 4B FP8 Scaled Text Encoder (CLIPLoader type: krea2)

Get

models/vae/ 253 MB HuggingFace

qwen_image_vae.safetensors Official 16-channel Video & Image VAE

Get

Field Notes RTX 4070 Benchmark

Benchmark: Rendered on NVIDIA GeForce RTX 4070 (12GB VRAM, CUDA 12.4)

1. CIVITAI REMIX & RESOURCE DECOMPOSITION: Ground-truth inspection of Civitai generation metadata for image #145657534 confirms an elegant, streamlined 2-component Krea 2 architecture: Sick Ollie FP8 base weights (Civitai Model #2676616, Version #3154408, File #3035266, AutoV2 hash FB51603FBB) combined with the character-specific Marla LoRA adapter (Civitai Model #2997563, Version #3399901, File #3290355, AutoV2 hash 1A6B2BDCE0, trigger 'dr0smarla') at strength 1.0. This combination delivers nuanced facial contours and rich nostalgic 35mm color grading. 2. WORKFLOW REUSE & REPRODUCTION STANDARD: Signal flow routes UNETLoader (sickOllie_krea2_3035266.safetensors, fp8_e4m3fn) and CLIPLoader (qwen3vl_4b_fp8_scaled.safetensors, type: krea2) sequentially through LoraLoader (dr0smarla_krea2_v1.safetensors). Diffusion runs via euler sampler paired with beta scheduler across 9 steps at CFG 1.0. Latents are decoded through the official 16-channel Qwen Image VAE (qwen_image_vae.safetensors), maintaining pure color balance without contrast invert artifacts. 3. HARDWARE BENCHMARKS & CONSUMER GPU FOOTPRINT: Benchmarked on NVIDIA GeForce RTX 4070 (12GB VRAM, CUDA 12.4). Peak VRAM stabilizes at 11.97 GB, operating comfortably within the 12GB threshold with zero layer offload swapping.

Frequently Asked Questions FAQ

What GPU and VRAM are required to run Civitai #145657534: Sick Ollie & Marla - Nostalgic Retro Lookbook (Krea 2)?

This workflow requires a minimum of 10GB VRAM (recommended 12GB VRAM). Tested and verified on NVIDIA GeForce RTX 4070 (12GB VRAM) at 1080x1440 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.