CivArchive
    Young Michael Scott Shaking Ed Trucks Hand Meme | Pose LoRA - v1.0
    NSFW
    Preview 3460730
    Preview 3460773
    Preview 3460729
    Preview 3460736

    Young Michael Scott Shaking Ed Trucks Hand Meme | Pose LoRA

    :D


    Weight can go from 0.9 up to 1. I recommend 1. Works with almost every Style, Concept and Photorealistic Models. Images were generated via Regional Prompt. Settings can be seen below. Reduce LoRA weight to 0.45-0.55, when using common prompt.

    Use of width 768 and height 512 recommended.

    Trigger Word:

    IncrsShakingHands

    Helping Prompt:

    handshake, standing, 2girl/2boys

    (there might be better settings, but I didn't see it. Switch between attention and latent)

    How to use LoRA's in auto1111:

    • Download LoRA

    • Copy the file to stable-diffusion-webui/models/lora

    • Select your LoRA and use the triggerword

    • Make sure to change the weight if needed (by default it's :1)

    • Have Fun!

    Description

    FAQ

    Comments (5)

    EnigmataNov 8, 2023
    CivitAI

    How to use regional prompter with Composable LoRA for applying loras by AND to each zone separately?

    firemanbrakeneckNov 8, 2023

    You don't need composable lora, regional's built in latent mode does that (I believe the original concept was copied from there). Unless it broke again.

    EnigmataNov 8, 2023

    @firemanbrakeneck and how it works? By using AND where first prompt before AND applies to all picture and other prompts apply to each part by their order? Or by using sections in extension?

    firemanbrakeneckNov 8, 2023

    Pretty much the same as attention: common clause is copied to all regions (including loras, in latent that's acceptable), base is an independent region which is applied to others with selected weight(s), and regions are applied according to divide ratio / placement of ADDROW ADDCOL etc.

    AND in latent mode is synonymous to BREAK (there's a checkbox for that which simply converts them in the prompt). The latent hooks all use region settings to setup the filter sizes and positions.

    Simple example: """ office <michaelscottlora> ADDCOMM <lora1> ADDCOL <lora2> """

    firemanbrakeneckNov 8, 2023· 1 reaction
    CivitAI

    @FallenIncursio The negative textencoder / unet settings are lora specific. Some corrupt, some get corrupted, and the values should be set per lora accordingly (through the regular comma separation like in divide ratio, just for loras instead of regions). For example, I think hu tao might be corrupting, based on how tartaglia's side looks.

    Refer to this section on lora corruption: https://github.com/hako-mikan/sd-webui-regional-prompter#known-issues

    LORA
    SD 1.5

    Details

    Downloads
    669
    Platform
    CivitAI
    Platform Status
    Available
    Created
    11/7/2023
    Updated
    5/15/2026
    Deleted
    -
    Trigger Words:
    IncrsShakingHands
    handshake, standing

    Files

    YoungMichaelScottShakingEdTrucksHand.safetensors