CivArchive
    Omnium - v1.0
    NSFW
    Preview 3491895
    Preview 3491898
    Preview 3491903
    Preview 3491904
    Preview 3491943
    Preview 3491902
    Preview 3491941
    Preview 3491901
    Preview 3491906
    Preview 3491905
    Preview 3491944
    Preview 3491912
    Preview 3491907
    Preview 3491909
    Preview 3491910
    Preview 3491946
    Preview 3491942
    Preview 3491911
    Preview 3491908

    OMNIUM

    Hereby I shared my merged model Omnium deriving it's name from the Latin word for everything, literally meaning of all. A suiting name since this merge contains a lot of models in it's DNA, and it's capable of quite a lot. It also hints at my intention to give back to the community as I am amazed by the stellar amount of quality output is shared within. This is part of my ongoing effort to contribute to this process.

    Listing all the models that went into Omnium would be an arduous task as I have been working on this for quite a while now and personally created over 164 merges to arrive at this point, that's more than 1 TB of data dedicated to this project. Even now I see room for improvement, but I decided it was time to share this with the community and see what you will do with it.

    Description

    FAQ

    Comments (9)

    lcmiracle22570Nov 9, 2023
    CivitAI

    Haven't given this a go yet, but have you already pruned the model or is this 6G reducable? Asking because I haven't being able to get my new rig with the 4060ti to work yet (something to do with Python environments and packages) so 6G is bit much for my 2080Ti atm.

    513820Nov 9, 2023

    This is the standard pruned size for XL models at this time. You may not have any luck running XL on a 2080, I run on a 3060 and it taxes my system to where I really can't even watch a 720 video with it running. Although, all things considered, I'm proud of my little old 3060 setup overall.

    lcmiracle22570Nov 9, 2023

    @MysteryWrecked Oh Ok, for some reason I thought this was 1.5... oh well, I will give it a go probably, if I can find that part of me willing to fix whatever is going on with my new rig...

    FrenzyX
    Author
    Nov 9, 2023· 1 reaction

    @lcmiracle22570 Basically what @MysteryWrecked said. However in one of my setups I run SDXL on a 2070S Mobile GPU, without any real issues, I use ComfyUI for this, and find it to work quite well. Nowadays there are also "SSD" models that have a smaller footprint. You could check out and try https://civitai.com/models/188863/segmind-ssd-1b

    shapeshifter83Nov 10, 2023

    ~6.5GB is SDXL standard. Your 4060Ti should be able to handle it fine unless you have the smaller VRAM version.

    GenXNov 9, 2023· 3 reactions
    CivitAI

    I love the color balance I'm able to get with this model. Great job!

    FrenzyX
    Author
    Nov 9, 2023

    That's nice, thank you!

    PrecisionMasterNov 14, 2023· 2 reactions
    CivitAI

    This model is Gold!
    It captures painting styles with great artistic emphasis!
    I'm stoked about what the next version will bring!

    FrenzyX
    Author
    Nov 15, 2023

    Thank you! Might be a while before a new version appears, but I have found myself planning for it already.
    If you could perhaps share some images that highlight what you mean, that would be appreciated, and might help me steer the model in the right direction in the future.

    Checkpoint
    SDXL 1.0

    Details

    Downloads
    881
    Platform
    CivitAI
    Platform Status
    Available
    Created
    11/9/2023
    Updated
    5/14/2026
    Deleted
    -

    Files

    omnium_v10.safetensors

    Mirrors

    HuggingFace (1 mirrors)
    CivitAI (1 mirrors)

    Available On (1 platform)

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