CivArchive
    Fefa Hentai Mix - V4
    NSFW
    Preview 1819041
    Preview 1227551
    Preview 1227535
    Preview 1227539
    Preview 1793531
    Preview 1227538
    Preview 1227542
    Preview 1227541
    Preview 1227544
    Preview 1227552
    Preview 1227560
    Preview 1227561
    Preview 1227563
    Preview 1227562
    Preview 1227564
    Preview 1802225
    Preview 1227534
    Preview 1227569
    Preview 1227568
    Preview 1227570

    FefaHentai | Checkpoint


    This is my hentai model for NSFW Hard generations without using LoRa.

    My other checkpoint FefaMix can generate good NSFW too, but works better using LoRa. If you want to work without porn's loras, I recommend this one

    Wanna Commission a LoRa or Get Early-Access? Go to my Patreon 💌

    Access my Patreon to have all my NSFW hard images, get a commission or learn about Stable Diffusion
    Your help will keep encouraging me to do more ❤️

    Recommendations of use 📋


    Sampling method: DPM++ 2M SDE Karras
    Sampling steps: 30
    Hires. fix: Yes
    Upscaler: I use R-ESRGAN 4x+ Anime6B, but you can try others
    Hires steps: 10-30
    Denoising: 0.3 - 0.5
    Resolution: 512x768 or higher
    Clip skip: 2

    ADetailer is highly recommended


    Which model choose?
    For SFW and less NSFW, you can check my other model FefaMix
    FefaMix has a good NSFW included and generates more 2.5D style

    For Hard NSFW and less SFW, you can use this one
    It generates more anime style than 2.5D.

    I use the FefaHentai model to generate Controlnet videos

    Help me buy a new setup 🖥️

    Donate to help me 🥺 Check out my Kofi ❤️

    Stay Connected: Join Me Online 💬

    Follow me on Instagram
    Follow me on Twitter
    Follow me on Patreon

    Like this model to receive updates 🥰

    Description

    If you have any questions related to the model or its samples, how to get started or just want to share your creations, feel free to join my discord: https://discord.gg/meinaverse
    ---------------------------------
    For this version have added more anime data and removed realistic data so it looks more anime than before.

    And i also adjusted the noise offset since i noticed in V3 the pictures were often rather too dark.

    FAQ