CivArchive
    LEOSAM'S FILMGIRL FILM STYLE LORA - NEW VELVIA VERSION - 2.0
    Preview 1
    Preview 2
    Preview 3
    Preview 4
    Preview 5
    Preview 6
    Preview 7
    Preview 8
    Preview 9
    Preview 10
    Preview 11
    Preview 12
    Preview 13

    Description

    • The training set image resolution of Velvia 2.0's main LoRa model was further increased to 768*1152.
      • When training the main model in this version, 15 film training photos of young Asian men were added to improve the model's realism when generating male photos.
      • On the basis of the high-resolution main model, the previous two editions of the Filmgirl historical model, which were well reviewed, were incorporated in a small proportion. After many tests, the operation of this fusion model can reduce the influence of the LoRa model on the face shape.

    FAQ

    LORA
    SD 1.5
    by 83a2e32a8f9345438a595f59002c3762
    Download (Beta) View on ShakkerAI
    Realistic
    Style & Filter
    Girl
    Style Boost
    Photography

    Details

    Downloads
    51,145
    Platform
    ShakkerAI
    Platform Status
    Available
    Created
    6/12/2023
    Updated
    9/27/2024
    Deleted
    -

    Files

    FilmVelvia2.safetensors

    Size:
    144.11 MB
    SHA256:
    803f3fa8d735fe885f8d6d0cb79f6fc1a033b18ab05e0dd020c050d7423ebda2

    Mirrors

    HuggingFace (39 mirrors)
    CivitAI (1 mirrors)
    TensorFiles (1 mirrors)
    ShakkerAI (1 mirrors)