CivArchive
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined
    Preview undefined

    ๐ŸŒŸ The Perfect Trio: Best Quantization Options

    1. ๐Ÿ”น Best Small Model: Q2_K

    • โšก Ultra-Fast inference speed

    • ๐Ÿ’พ Tiny Size: 8x smaller than original

    • ๐Ÿ’ป Perfect for low-resource devices

    • ๐Ÿ”‹ Ideal when speed > perfect accuracy

    2. ๐Ÿ”ธ Best All-Rounder: Q4_K_M

    • โš–๏ธ Perfect Balance: size vs quality

    • ๐Ÿง  Strong reasoning capabilities

    • ๐Ÿ‘‘ Community Favorite for daily use

    • ๐ŸŽฏ Go-to choice for most applications!

    3. ๐Ÿ”ท Premium Quality: Q8

    • โœจ Nearly Identical to original model

    • ๐Ÿงฉ Preserves complex reasoning abilities

    • ๐ŸŽจ Superior creative generation

    • ๐Ÿ’ช Best when quality is non-negotiable

    ๐Ÿ› ๏ธ Complete Installation Guide

    ๐Ÿ“ Setup Structure

    ๐Ÿ“‚ ComfyUI/
    โ”œโ”€โ”€ ๐Ÿ“‚ models/
    โ”‚   โ”œโ”€โ”€ ๐Ÿ“‚ diffusion_models/
    โ”‚   โ”‚   โ”œโ”€โ”€ (basic)๐Ÿ“„ SPEED_Q8.gguf 
    โ”‚   โ”œโ”€โ”€ ๐Ÿ“‚ text_encoders/
    โ”‚   โ”‚   โ”œโ”€โ”€ (basic)๐Ÿ“„ clip_l.safetensors
    โ”‚   โ”‚   โ”œโ”€โ”€ (option1)๐Ÿ“„ t5xxl_fp16.safetensors
    โ”‚   โ”‚   โ”œโ”€โ”€ (option2)๐Ÿ“„ t5xxl_fp8_e4m3fn.safetensors
    โ”‚   โ”‚   โ””โ”€โ”€ (option3)๐Ÿ“„ t5xxl_fp8_e4m3fn_scaled.safetensors
    โ”‚   โ”œโ”€โ”€ ๐Ÿ“‚ vae/
    โ”‚   โ”‚   โ””โ”€โ”€ ๐Ÿ“„ ae.safetensors

    ๐Ÿ’Ž Essential Components

    This merged model offers a balanced solution for AI-driven image generation, emphasizing both speed and quality. Whether you're processing single images or large batches, it delivers high-quality visuals efficiently.

    ๐Ÿ”ค Text Encoders - The Brain Behind Natural Language Understanding

    Note: You only need to choose ONE of the T5XXL options below based on your hardware capabilities

    ๐ŸŽญ VAE - The Visual Artist

    ๐Ÿ”ฎ Pro Workflows

    ๐Ÿ‘ Special Thanks

    Big thanks to city96 for pioneering the GGUF journey! ๐Ÿ™Œ

    ๐Ÿ‘จโ€๐Ÿ’ป Developer Information

    This workflow guide was created by Abdallah Al-Swaiti:

    For additional tools and updates, check out the OllamaGemini Node: GitHub Repository

    No alternative text description for this image

    Description

    FAQ

    Comments (7)

    marinosakuraAug 19, 2024
    CivitAI

    how to get the best result with this model with Forge? step? scheduler?, CFG scale etc?

    AbdallahAlswa80
    Author
    Aug 19, 2024

    i didn't test it on forge ,, whatever on comfyui , 4-8 step is enough , cfg =1 , for upscaler 2 - 3 enough

    Diana111Aug 19, 2024
    CivitAI

    Will there be a version without F1.dev?

    AbdallahAlswa80
    Author
    Aug 19, 2024

    the base here scnell

    Diana111Aug 19, 2024

    @AbdallahAlswa80ย But with the addition of Dev, the license becomes non-commercial.
    Separately, schnell permits commercial use.

    AbdallahAlswa80
    Author
    Aug 19, 2024

    @Diana111ย  realy i dont know may this need to contact black forest !

    Diana111Aug 20, 2024

    @AbdallahAlswa80 They definitely wrote it in the license (for checkpoint based on Dev):
    Non-Commercial Use Only. You may only access, use, Distribute, or creative Derivatives of or the FLUX.1 [dev] Model or Derivatives for Non-Commercial Purposes.

    I really liked your previous model, so I asked if I should wait for a new one that I can use.

    Checkpoint
    Flux.1 S

    Details

    Downloads
    611
    Platform
    CivitAI
    Platform Status
    Available
    Created
    8/19/2024
    Updated
    5/13/2026
    Deleted
    -

    Files

    speedQ8_ggufQ8SCHNELLDEV.zip

    Mirrors

    HuggingFace (1 mirrors)
    CivitAI (1 mirrors)

    Available On (1 platform)

    Same model published on other platforms. May have additional downloads or version variants.