CivArchive
    Preview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefinedPreview undefined

    Note: The model have moved completely to Tensor

    you can still download it there :

    https://tensor.art/models/642019060365894429

    A new dawn has come to this place!

    This is a general model it can handel a wide range of different images

    • Photorealistic

    • Digital Art

    • Anime

    • Comic

    • Vektor Grapics

    • Line art

    • Stickers

    • can write short text

    • etc

    The Model needs no refiner or any enhancement lora

    V1.6

    is my first attempt to create a SD 1.0xl model.

    I used as base :

    XL3 - Experimental SD1.0 XL

    https://civarchive.com/models/118312?modelVersionId=128319

    this is a (WIP) Model

    V2.2

    complete overhaul of the old version

    merged with the newest versions of FenrisXL and XL6

    and did also get some new training on top

    But still WIP

    V3.1

    Merged in parts of FenrisXL 8.09 and 9.22

    i also did add some training

    • more creative

    • better image quality

    • more consistend than v2.2

    PS: will report every sick photo (that is against the law)

    and also please check out FenrisXL

    made by Freek22

    Description

    this is the bf16 version of newdawnxl because it has a bit other results than the fp16 Version iseperated both versions.

    • recommended for NVIDIA cards

    otherwise use the fp16 version!