CivArchive
    Preview 149561521
    Preview 149561546

    Ultra Realistic by Stable Yogi · Krea 2 · V1.5

    Pictures that look like a camera took them, now with a more natural, everyday phone-photo feel. Richer skin with visible texture and small marks, and lighting that behaves like it does in a real room.

    Download the Pro version from my portal

    🆕 What is new in V1.5

    • A more natural amateur, phone-photo look. Less polished, more real.

    • Richer skin detail and texture you can see at 1:1.

    • Better lighting realism: window light, flash, golden hour, night street.

    • Everything you liked in the first version, built on top of it.

    🎯 Great for

    • Photo skin with pores and fine texture, not smooth plastic.

    • Casual snapshot looks, like a picture taken with a phone.

    • Strong bodies and poses. Adult anatomy that holds together.

    • Speed. A finished picture in 12 steps.

    💾 One file per version. Pick by your card

    • 16 GB and up: INT8-ConvRot

    • 16 GB and up: fp8-scaled

    • 24 GB and up: Q8_0 GGUF

    • 8 GB and up: NVFP4 (RTX 50-series only)

    • 8 GB and up: INT4-ConvRot (RTX 20-series and up)

    • 8-12 GB: Q4_0 GGUF

    • 6 GB: Q2_K GGUF Q3 GGUF

    Leave about 2 GB spare for the text encoder. A file that is too big still runs, only slower. GGUF files need the ComfyUI-GGUF nodes. On AMD, use a GGUF file. Pair the model with the Qwen3-VL text encoder and the Qwen-Image VAE, the same as base Krea 2.

    ⚙️ Settings
    12 steps · CFG 1.5 · ER SDE · Beta · 1024×1536, or any size near 1.5 megapixels. These are the settings used for every sample picture.

    ⭐ Want more punch? The Pro edition has richer detail and more presence.

    Download the Pro version from my portal

    Made by Stable Yogi 🧘. Drop a ❤️ and share what you make!

    ====================================================

    Ultra Realistic by Stable Yogi · Krea 2 · V1.0

    Pictures that look like a camera took them. Real skin with pores and small marks, light that falls like it does in a real room, bodies that look like real bodies. Pro version on my portal.

    🎯 Great for

    • Photo skin you can see at 1:1, not smooth plastic.

    • Natural light: window light, flash, golden hour, night street.

    • Strong bodies and poses. Adult anatomy that holds together.

    • Detail. A finished picture in 12 steps.

    💾 One file per version. Pick by your card

    • 16 GB and up: INT8-ConvRot (13.5 GB)

    • 16 GB and up: fp8-scaled (13.1 GB)

    • 24 GB and up: Q8_0 GGUF (14.6 GB)

    • 8 GB and up: NVFP4 (7.7 GB)

    • 8 GB and up: INT4-ConvRot (7.4 GB)

    • 8-12 GB: Q4_0 GGUF (8.3 GB)

    • 6 GB: Q2_K GGUF (5.4 GB)

    Leave about 2 GB spare for the text encoder. A file that is too big still runs, only slower. GGUF files need the ComfyUI-GGUF nodes. On AMD, use a GGUF file. Pair the model with the Qwen3-VL text encoder and the Qwen-Image VAE, the same as base Krea 2.

    ⚙️ Settings
    12 steps · CFG 1.5 · ER SDE · Beta · 1024×1536, or any size near 1.5 megapixels.

    ⭐ Want more punch? The PRO edition has richer detail and more presence.

    Made by Stable Yogi 🧘. Drop a ❤️ and share what you make!

    Description

    Ultra Realistic by Stable Yogi · Krea 2 · V1.5 · Q3_K GGUF

    Small file for 8 GB cards. Sharper than Q2_K, lighter than Q4_0. Needs the ComfyUI-GGUF nodes.

    File size: 6.7 GB. Card: 8 GB.

    Settings: 12 steps · CFG 1.5 · ER SDE · Beta · 1024×1536. Use the Qwen3-VL text encoder and the Qwen-Image VAE.

    Download the Pro version from my portal

    FAQ

    Checkpoint
    Krea 2

    Details

    Downloads
    573
    Platform
    CivitAI
    Platform Status
    Available
    Created
    11/20/2026
    Updated
    11/22/2026
    Deleted
    -

    Files

    Ultra Realistic by Stable Yogi Krea2 V1.5 Civ Q3_K C95.gguf

    Ultra Realistic by Stable Yogi Krea2 V1.5 Civ Q3_K C95.gguf