CivArchive
    [Community] LoRA extract - LORADM320realisticVisionV13_v13
    NSFW
    Preview 292874
    Preview 292873
    Preview 292872

    LoRA extractions is not as good as the full model but it creates a compromise between size and quality and it can be used along other LoRA model with a full model. Credit goes the creator of the checkpoint in version description

    Sorry, i do not accept anymore request link(s) in my comments "Suggestions for a model extraction?"

    04.0417.23 Name version is limited to 10 25 character, make it impossible to see full name... see below "List of all LoRA model".

    How to use it in AUTOMATIC1111 WebUI?

    • Place the LoRA model in .\stable-diffusion-webui\models\Lora then use <lora:filename:multiplier> in the prompt.

    About the comparison:

    • Picture order: First = LoRA, Second = Fine-tuned Model, Third = Base Model

    • DM = Network Dimension ( 120=lower quality/low size, 320=higher quality/higher size ) same for CD = Conv Dimension (Rank)
      [CD is optionally if you wish to use it, install the extension: https://github.com/KohakuBlueleaf/a1111-sd-webui-locon (How? Extensions -> Install from URL)]
      Obviously Version "1024CV1024" or in short "1024" use more VRAM
      I did a large comparison DM+CV here

    • sample pictures does not use vea and LoRA model with SDv1.4

    • every model from (Stable Diffusion Base Model: mostly sd-v1-4.ckpt, model base is determined by highest % affinity) was extracted using Kohya_ss (Save precision: fp16)

    • Name format: <Type><DM> <CV><Name>

    List of all LoRA model:

    Description

    base: realisticVisionV13_v13.safetensors [de2f2560]

    v2-1_768-ema-pruned.ckpt [4bdfc29c] - -0.29% v1-5-pruned.ckpt [a9263745] - 78.82%

    sd-v1-4.ckpt [7460a6fa] - 88.22%

    FAQ

    Comments (2)

    GlennDaveMar 27, 2023· 1 reaction
    CivitAI

    Excellent work and thank you for submitting for the rest of the world to view. You are correct that the Lora falls somewhere between, but I was hoping that it would be closer to the fine tuned model versus the 1.4 checkpoint. I wonder if there are methods to get it even closer to the fine tuned model or if this is just the state of the art as it stands now.

    DigitalDreamer
    Author
    Mar 29, 2023

    @GDC I already train my BerryGold dataset on new LoRA (not the extract one), this sometimes give me better result with a large DM LoRA (keep in mind that is not just one concept)...
    Also "LORA1024CD1024perfectWorld_v2Baked" LoRA of 2GB (DM1024 + CV1024) that is a little better but probably not worth size trade as a pruned full model have the same size.

    LORA
    SD 1.4

    Details

    Downloads
    282
    Platform
    CivitAI
    Platform Status
    Available
    Created
    3/21/2023
    Updated
    5/14/2026
    Deleted
    -

    Files

    LORADM320realisticVisionV13_v13.safetensors

    Available On (1 platform)

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