⚠️ License & Usage Disclaimers

Please review the full license agreement before accessing or using the models.

The correct license and permissions can be found at:
https://huggingface.co/FFusion/FFXL400/blob/main/LICENSE.md

"Model Weights: The weights used for the models/loras are provided "as is." FFusion AI and Source Code Bulgaria do not grant any rights for their commercial use. These weights are strictly for testing and experimental purposes.



ORIGIN OF LORAS:  



The LORAs and weights provided are extracted from SDXL models (checkpoints). 



All licenses, terms, and conditions set forth by the original checkpoint creator must be respected and adhered to."

❤️ Acknowledgments to Model Creators

We utilized the following checkpoints in the creation of this merged model:

UPDATE: Checkpoint Version 2.0pre

EXTRACTED Weights (as LoRA's) from:

Model.0003.4Guofeng4xl_V1125d

Model.0004.Ambiencesdxl_A1

Model.0006.Angrasdxl10_V22

Model.0007.Animaginexl_V10

Model.0008.Animeartdiffusionxl_Alpha3

Model.0009.Astreapixiexlanime_V16

Model.0010.Bluepencilxl_V010

Model.0011.Bluepencilxl_V021

Model.0012.Breakdomainxl_V03d

Model.0013.Canvasxl_Bfloat16v002

Model.0014.Cherrypickerxl_V20

Model.0015.Copaxtimelessxlsdxl1_V44

Model.0016.Counterfeitxl-Ffusionai-Alpha-Vae

Model.0017.Counterfeitxl_V10

Model.0018.Crystalclearxl_Ccxl

Model.0019.Deepbluexl_V006

Model.0020.Dream-Ffusion-Shaper

Model.0021.Dreamshaperxl10_Alpha2xl10

Model.0022.Duchaitenaiartsdxl_V10

Model.0023.Dynavisionxlallinonestylized_Beta0371bakedvae

Model.0024.Dynavisionxlallinonestylized_Beta0411bakedvae

Model.0025.Fantasticcharacters_V55

Model.0026.Fenrisxl_V55

Model.0027.Fudukimix_V10

Model.0028.Infinianimexl_V16

Model.0029.Juggernautxl_Version1.lora_1.safetensors

Model.0030.Lahmysterioussdxl_V330

Model.0031.Mbbxlultimate_V10rc

Model.0032.Miamodelsfwnsfwsdxl_V30

Model.0033.Morphxl_V10

Model.0034.Nightvisionxlphotorealisticportrait_Beta0681bakedvae.lora_1.safetensors

Model.0035.Osorubeshialphaxl_Z

Model.0036.Physiogenxl_V04

Model.0037.Protovisionxlhighfidelity3d_Beta0520bakedvae

Model.0038.Realitycheckxl_Alpha11

Model.0039.Realmixxl_V10

Model.0040.Reproductionsdxl_V31

Model.0041.Rundiffusionxl_Beta

Model.0042.Samaritan3dcartoon_V40sdxl

Model.0043.Sdvn6realxl_Detailface

Model.0044.Sdvn7realartxl_Beta2

Model.0045.Sdxl10arienmixxlasian_V10

Model.0046.Sdxlbasensfwfaces_Sdxlnsfwfaces03

Model.0047.Sdxlfaetastic_V10

Model.0048.Sdxlfixedvaefp16remove_Basefxiedvaev2fp16

Model.0049.Sdxlnijiv4_Sdxlnijiv4

Model.0050.Sdxlronghua_V11

Model.0051.Sdxlunstablediffusers_V5unchainedslayer

Model.0052.Sdxlyamersanimeultra_Yamersanimev2

Model.0053.Shikianimexl_V10

Model.0054.Spectrumblendx_V10

Model.0055.Stablediffusionxl_V30

Model.0056.Talmendoxlsdxl_V11beta

Model.0057.Wizard_V10

Model.0058.Wyvernmix15xl_Xlv11

Model.0059.Xl13asmodeussfwnsfw_V17bakedvae

Model.0060.Xl3experimentalsd10xl_V10

Model.0061.Xl6hephaistossd10xlsfw_V21bakedvaefp16fix

Model.0062.Xlperfectdesign_V2ultimateartwork

Model.0063.Xlyamersrealistic_V3

Model.0064.Xxmix9realisticsdxl_Testv20

Model.0065.Zavychromaxl_B2

Model.0066.hassakuSfwNsfwAlphav_alphaV02

Model.0067.galaxytimemachinesGTM_xlplusV10

Model.0068.furtasticxl_BetaEPOCHS3

Model.0069.formulaxlXLComfyui_v20Pruned

Model.0070.FinalAnimeCG_mk2a2

Model.0071.explicitFreedomNSFW_beta

Model.0072.endjourneyXL_v11

Model.0073.dreamshaperXL10_alpha2Xl10

Model.0074.copaxTimelessxlSDXL1_v5

Model.0075.cinemaxAlphaSDXLCinema_alpha1

Model.0076.brixlAMustInYour_v20Banu

Model.0077.animeChangefulXL_v10ReleasedCandidate

Model.0078.xlYamersCartoonArcadia_v1

Model.0079.venusxl_v11

Model.0080.unsafexl_v20

Model.0081.sdxlYamersRealism_version2

Model.0082.sdxlUnstableDiffusers_v6StabilityEater

Model.0083.sdxlNuclearGeneralPurposeSemi_v10

Model.0084.sdvn6Realxl_detailface

Model.0085.samaritan3dCartoon_v40SDXL

Model.0086.realvisxlV10_v10VAE

Model.0087.RealitiesEdgeXLANIME_20

Model.0088.RealitiesEdgeXL_30

Model.0089.realisticStockPhoto_v10

Model.0090.realisticFreedomSFW_alpha

Model.0091.realcartoonXL_v2

Model.0092.pyrosSDModelsBlowjob_v0122022steps

Model.0093.pyrosNSFWSDXL_v013e6

Model.0094.nightvisionXLPhotorealisticPortrait_v0743ReleaseBakedvae

Model.0095.newone_v10

Model.0096.MOHAWK_v10BETA

Model.0097.juggernautXL_version4

Model.0098.sdxlYamersRealism_version2

? The models and weights available in this repository are strictly for research and testing purposes, with exceptions noted below. They are not generally intended for commercial use and are dependent on each individual LORA.

? Exception for Commercial Use: The FFusionXL-BASE, FFusion-BaSE, di.FFUSION.ai-v2.1-768-BaSE-alpha, and di.ffusion.ai.Beta512 models are trained by FFusion AI using images for which we hold licenses. Users are advised to primarily use these models for a safer experience. These particular models are allowed for commercial use.

? Disclaimer: FFusion AI, in conjunction with Source Code Bulgaria Ltd and BlackswanTechnologies, does not endorse or guarantee the content produced by the weights in each LORA. There's potential for generating NSFW or offensive content. Collectively, we expressly disclaim responsibility for the outcomes and content produced by these weights.

? Acknowledgement: The FFusionXL-BASE model model is a uniquely developed version by FFusion AI. Rights to this and associated modifications belong to FFusion AI and Source Code Bulgaria Ltd. Ensure adherence to both this license and any conditions set by Stability AI Ltd for referenced models.

FFXL400(GB) Combined LoRA Model ?
400GB-LoraXL repository. Our vision was to harness the power of multiple LoRAs, meticulously analyzing and integrating a select fraction of the blocks from each.

? Model Highlights

• Innovative Combination: This model is a strategic integration of LoRAs, maximizing the potential of each while creating a unified powerhouse.

• Versatility: The model is available in various formats including diffusers, safetensors (both fp 16 and 32), and an optimized ONNIX FP16 version for DirectML, ensuring compatibility across AMD, Intel, Nvidia, and more.

• Advanced Research: Leveraging the latest in machine learning research, the model represents a state-of-the-art amalgamation of LoRAs, optimized for performance and accuracy.

? Technical Insights

This model is a testament to the advancements in the field of AI and machine learning. It was crafted with precision, ensuring that:

• Only a small percentage of the blocks from the original LoRAs (UNet and text encoders) were utilized.

• The model is primed not just for inference but also for further training and refinement.

• It serves as a benchmark for testing and understanding the cumulative impact of multiple LoRAs when used in concert.

? Usage

The FFXL400 model is designed for a multitude of applications. Whether you're delving into research, embarking on a new project, or simply experimenting, this model serves as a robust foundation. Use it to:

• Investigate the cumulative effects of merging multiple LoRAs.

• Dive deep into weighting experiments with multiple LoRAs.

• Explore the nuances and intricacies of integrated LoRAs.

? How to Use

The model can be easily integrated into your projects. Here's a quick guide on how to use the FFXL400 model:

1. Loading the Model:

   from transformers import AutoModel, AutoTokenizer
   
   
   
   tokenizer = AutoTokenizer.from_pretrained("FFusion/FFXL400")
   
   model = AutoModel.from_pretrained("FFusion/FFXL400")
   
   

2. Performing Inference:

   input_text = "Your input here"
   
   inputs = tokenizer(input_text, return_tensors='pt')
   
   with torch.no_grad():
   
       outputs = model(**inputs)
   
   

Further Training

You can also use the FFXL400 as a starting point for further training. Simply load it into your training pipeline and proceed as you would with any other model.

Autotrain Advanced,

Kohya + Stable Diffusion XL,

? Background

The FFXL400 is built upon the insights and data from the 400GB-LoraXL repository. Each LoRA in that collection was extracted using the Low-Rank Adaptation (LoRA) technique, providing a rich dataset for research and exploration. The FFXL400 is the pinnacle of that research, representing a harmonious blend of the best LoRAs.

Library of Available LoRA Models ?

? Acknowledgements & Citations

A huge shoutout to the community for their continued support and feedback. Together, we are pushing the boundaries of what's possible with machine learning!

pip install howmuch

howmuch

howmuch --scan C: