CivArchive
    AnRealSpiceMix - v1.0
    NSFW
    Preview 3734598
    Preview 3734599
    Preview 3734601
    Preview 3734600
    Preview 3734724
    Preview 3734821

    Hello everyone,

    I experimented merging some of my models. I liked the results. So I thought why not publishing it. Let me know if you like it or if its unnecessary.

    This is a merge from my other models: AnReal, realspice and RealLife.

    V1.0: AnReal, realspice and RealLife

    V2.0: AnReal, realspice, RealLife and RealHotSpice

    V3.0: AnReal, realspice, RealLife, RealHotSpice and WorldOfFusch

    I have used these settings almost exclusively. They are a good starting point and can be adjusted depending on what you want to generate.

    Sampling method:

    • DPM++ 2M Karras

    • DPM++ 2M SDE Karras

    Sampling Steps:

    • 25-45

    • CFG Scale: 5-7

    • Clip Skip: 1-2

    Recommended Upscaler Settings:

    Hires. fix Upscaler:

    • Hires steps: 10-20

    • Hires upscale: 1.5-2

    • Denoising strength: ~ 0.25-0.5

    If you can use Hires.fix for better Results!

    I don't use restore faces. If you got a bad face or want to change it use ADetailer for it!

    I would greatly appreciate any feedback!

    If you like this model, I would greatly appreciate a positive review along with constructive criticism, suggestions for improvement, or anything similar!

    Description

    FAQ

    Comments (6)

    mayzonuzumaki839Dec 3, 2023
    CivitAI

    What's the difference between the 4GB and 2GB versions?

    davzDec 5, 2023· 3 reactions

    The 2GB is the pruned model, which is the one you want. You only want the full (unpruned) model if you're looking to use it as a base for further training.

    32BitshifterDec 6, 2023· 10 reactions

    about 2gb

    ralphywDec 9, 2023· 4 reactions

    @32Bitshifter Dad, how did you get back on the computer?

    templeam711Feb 16, 2024

    @davz Can you help me understand how much difference there is between an fp16 and an fp32 model? The advice I have read about pruned vs full always says what you say here, but I have not found a good comparison of fp16 vs fp 32. Is it better image quality? More knowledge of language?

    davzFeb 16, 2024· 2 reactions

    @templeam711 fp16 is smaller in file size (and VRAM), fp32 is higher precision. For image generation in theory fp32 should offer better quality. But in practice the difference is hardly noticeable. If you're training or merging models though that extra precision might be worth the extra VRAM and time. I've never trained or merged models myself, so you'd need to ask someone else if that's what you're looking to do.

    Checkpoint
    SD 1.5

    Details

    Downloads
    2,547
    Platform
    CivitAI
    Platform Status
    Available
    Created
    11/19/2023
    Updated
    4/30/2026
    Deleted
    -

    Files

    anrealspicemix_v10.safetensors

    Mirrors

    Available On (2 platforms)

    Same model published on other platforms. May have additional downloads or version variants.