CivArchive
    FLUX.1 T5 Text Encoder - Stock
    Preview 28532744

    This is the T5 text encoder that is directly uploaded from HuggingFace here.

    https://huggingface.co/comfyanonymous/flux_text_encoders/tree/main

    These have been uploaded previously but they used a NSFW flagged image for the thumbnail making it inaccessible for users on CivitAI Green.

    This was uploaded for the purpose of Chromafur but you can use it with any FLUX.1 model.

    fp16: t5xxl_fp16.safetensors

    fp8 t5xxl_fp8_e4m3fn.safetensors

    Description

    FAQ

    Comments (2)

    TomLucidorOct 19, 2024
    CivitAI

    Here is a Q: at which quantization would it be good for those with 12GB or even 8GB of VRAM, without compromising on quality?

    ranger89Apr 19, 2025
    CivitAI

    On the map 2060Super my generation time is 1 min 35 sek.Everything works fine!

    Checkpoint
    Flux.1 D
    by HorizonTeam
    Download (Beta) View on CivitAI
    base model
    flux.1
    flux1.d
    flux1.s
    chromafur
    t5xxl
    textencoder

    Details

    Downloads
    2,935
    Platform
    CivitAI
    Platform Status
    Available
    Created
    9/9/2024
    Updated
    5/13/2026
    Deleted
    -

    Files

    flux1T5TextEncoder_stock.safetensors

    Size:
    4.56 GB
    SHA256:
    7d330da4816157540d6bb7838bf63a0f02f573fc48ca4d8de34bb0cbfd514f09

    Mirrors

    HuggingFace (167 mirrors)
    CivitAI (4 mirrors)
    CivitasBay (1 mirrors)
    ModelScope (1 mirrors)
    ModelScope CN (3 mirrors)

    flux1T5TextEncoder_stock.safetensors

    Size:
    9.12 GB
    SHA256:
    6e480b09fae049a72d2a8c5fbccb8d3e92febeb233bbe9dfe7256958a9167635

    Mirrors

    HuggingFace (174 mirrors)
    CivitAI (5 mirrors)
    CivitasBay (1 mirrors)
    ModelScope (2 mirrors)
    ModelScope CN (2 mirrors)
    ShakkerAI (1 mirrors)