CivArchive
    Repost sd3-t5xxl-fp16 model - t5xxl-fp16
    Preview 1

    Description

    Used in conjunction with sd3 workflow large models

    1. Update comfyui to the latest version, which is version c8b5e08

    2. Import the json workflow into comfyui, or drag it directly into comfyui

    3. Download all text encoders, and at least one large model

    That's right, these are the text encoders, fp8 and fp16, choose one, if the configuration is not enough, you may need fp8

    4. Put the text encoder in the clip folder under the models folder in comfyui

    5. Put the large model in the checkpoints folder under the models folder in comfyui

    You can download all four text encoders completely on my homepage, for other large models, you need to go to the official account to find them

    Below is the link for clip-g and clip-l text encoders

    https://www.liblib.art/modelinfo/d145cade95b44fcb9b3c317836f1a1c2

    Other
    SDXL
    by 6163075250364aa39c3149db9c8eac09
    Download (Beta) View on ShakkerAI
    Realistic
    Portrait Photography
    Boy
    Physical/Facial feature
    Photography

    Details

    Downloads
    13
    Platform
    ShakkerAI
    Platform Status
    Available
    Created
    6/13/2024
    Updated
    6/13/2024
    Deleted
    -

    Files

    t5xxl_fp16.safetensors

    Size:
    9.12 GB
    SHA256:
    6e480b09fae049a72d2a8c5fbccb8d3e92febeb233bbe9dfe7256958a9167635

    Mirrors

    HuggingFace (178 mirrors)
    CivitAI (5 mirrors)
    CivitasBay (1 mirrors)
    ModelScope (2 mirrors)
    ModelScope CN (2 mirrors)
    ShakkerAI (1 mirrors)