🌟 Trained specifically for style mixing on Safe For Work (SFW) dataset.
🔹 Text Encoder (strongly reliant on U-Net architecture):
🔹 Lora Model:
Trained using the CivitAI trainer.
Underwent training for 16 epochs.
Performance: Lora showcases stable diffusion in XL settings.
🔹 Lycoris Models:
Configured using this preset from LyCORIS, but with our custom tweaks.
Trained on 8xA6000 GPUs.
Underwent 20-40 epochs with a batch size of 20.
Module Type Breakdown:
LohaModule: 176
LoConModule: 150
FullModule: 26
LokrModule: 700
🏆 Comparison:
CivitAI's Lora stands out in our tests, but we've yet to explore its full potential in style mixing with other models.
Feedback, collaboration, and tests are welcome! 🤟 🥃
Description
Text Encoder Only
{
"ssmd_author": "FFusion AI",
"modelspec.prediction_type": "epsilon",
"modelspec.title": "FFelf-LiFFe-FFusion-LoRA-FA-TX10_v2",
"ss_base_model_version": "sdxl_base_v1-0",
"modelspec.date": "2023-09-23T03:44:50",
"ss_network_alpha": "4.0",
"ssmd_tags": "",
"ss_network_module": "networks.lora_fa",
"ssmd_description": "🌐 Contact Information\n\nThe FFusion.ai project is proudly maintained by Source Code Bulgaria Ltd & Black Swan Technologies.\n\n📧 Reach us at [email protected] for any inquiries or support.\n\n🌌 Find us on:\n\n🐙 GitHub\nhttps://github.com/1e-2\n\n😊 Hugging Face\nhttps://huggingface.co/ffusion\n\n💡 Civitai\nhttps://civitai.com/user/idle/models\n\n\n🌍 Sofia Istanbul London\n\nhttps://FFusion.ai\nhttps://FFAI.eu\nhttps://1e-2.com",
"ss_v2": "False",
"modelspec.implementation": "https://github.com/Stability-AI/generative-models",
"modelspec.resolution": "1024x1024",
"modelspec.architecture": "stable-diffusion-xl-v1-base/lora",
"sshs_model_hash": "3284f245a0967202532fcfccd58b6ac90f13ca079f2a9cefec28cf6d2732a0a5",
"ssmd_display_name": "FF Safe AnimXL (Text Encoder)",
"ssmd_source": "https://ffusion.ai/",
"ss_network_dim": "4",
"modelspec.encoder_layer": "2",
"modelspec.sai_model_spec": "1.0.0",
"sshs_legacy_hash": "04567737",
"ssmd_keywords": "",
"ssmd_rating": "0"
}FAQ
Comments (1)
SAFE Anime TX encoder actually works well to make over trained models generate anime style, such as Juggernaut XL or copxTimelessXL, it is very useful, otherwise these models would only generate realistic photos.