CivArchive
    Cloudify - SDXL v1.1
    Preview 1851863Preview 1851879Preview 1851901Preview 1851979Preview 1852091Preview 1852143Preview 1852283Preview 1852345Preview 1852415Preview 1852585Preview 1851982

    Choose SDXL (LoRA) or SD 1.5 version from the menu on the top. Rate, like, and review to help people find the model on the site.

    Cloudify is a tool for generating cloudy skies. It was trained on a dataset of my photos of the sky during the day.

    SDXL: Trained on 1024x1024 from hi-res photos. The dataset is updated with more skies and effects. For the SDXL version, use weights 0.5-1.2, go higher for texturing depending on your prompt.

    SD1.5: This LyCORIS/LoHA experiment was trained on 512x512 from hires photos, so I suggest upscaling it from there (it will work on higher resolutions directly, but it seems to override other subjects more frequently).

    • Use low weights for misty effects (0.1+)

    • Use high weights for cloud textures (0.8+)

    • Use Hires. fix (4x_RealisticRescaler, 4x-Ultrasharp upscalers with denoising strength around 0.3-0.6) for best details in txt2img

    • Check the gallery for ideas, on how to incorporate other Loras

    If you want to learn more about AI-augmented design and algorithmic art, check my web pages.

    For my notes on training and use of stable diffusion models, check https://sandner.art/articles/ai/

    If you can afford it, support my work by buying me a coffee to keep those costly render cores go: https://ko-fi.com/airesearch

    Review or rate the model, it helps.

    Have fun rendering!

    Description

    The dataset expanded from v1.0 (new skies, daytime, test rainbow), training time 5h 2000s. 1024x1024

    LORA
    SDXL 1.0

    Details

    Downloads
    2,063
    Platform
    CivitAI
    Platform Status
    Available
    Created
    8/3/2023
    Updated
    9/27/2025
    Deleted
    -
    Trigger Words:
    cloudify
    sky
    clouds
    rainbow

    Files

    Cloudify-SDXL.safetensors

    Mirrors

    Huggingface (1 mirrors)
    CivitAI (1 mirrors)