Pony Diffusion V6 XL

提供商LyliaEngine
分类text-to-image
许可证cdla-permissive-2.0
下载量242.2K
星标0

简介

Pony Diffusion V6 XL 是基于 SDXL 微调的图像生成模型，在二次元和动漫风格领域具有极强的统治力。它最核心的突破在于对自然语言指令的深度理解，用户不再需要堆砌冗长的提示词标签，通过简单的描述即可精准控制角色姿势和场景。对于习惯了 Stable Diffusion 的开发者和创作者来说，它在构图准确度和画风纯净度上显著优于原版 SDXL，是目前制作高质量动漫插画、角色同人图的首选模型，上手门槛低且出图效率极高。

核心亮点

  • 极强的自然语言理解力，无需复杂 Tag 堆砌
  • 动漫与二次元风格顶尖，角色还原度极高
  • 基于 SDXL 架构，兼容主流 WebUI 插件
  • 精准控制人物姿态与复杂场景构图

使用方法

安装依赖
# 安装 Hugging Face transformers
pip install transformers torch
SDK 使用
# 使用 transformers 加载模型
from transformers import AutoModel, AutoTokenizer

model = AutoModel.from_pretrained("LyliaEngine/Pony_Diffusion_V6_XL")
tokenizer = AutoTokenizer.from_pretrained("LyliaEngine/Pony_Diffusion_V6_XL")

Hugging Face 下载

我们推荐使用命令行或者 Hugging Face Hub SDK 来进行模型的下载。

操作指引：在下载前，请先通过如下命令安装 huggingface_hub：

操作指引 
pip install -U huggingface_hub

命令行下载

下载完整模型库

下载完整模型库 
huggingface-cli download LyliaEngine/Pony_Diffusion_V6_XL

下载单个文件到指定本地文件夹（以下载 config.json 到当前路径下 ./dir 目录为例）

下载单个文件到指定本地文件夹（以下载 config.json 到当前路径下 ./dir 目录为例） 
huggingface-cli download LyliaEngine/Pony_Diffusion_V6_XL config.json --local-dir ./dir

更多命令行下载选项，可参见官方文档

SDK 下载

SDK 下载 
# 模型下载
from huggingface_hub import snapshot_download
model_dir = snapshot_download('LyliaEngine/Pony_Diffusion_V6_XL')

Git 下载

请确保 lfs 已经被正确安装

Git 下载 
git lfs install
git clone https://huggingface.co/LyliaEngine/Pony_Diffusion_V6_XL

如果您希望跳过 lfs 大文件下载，可以使用如下命令

跳过 LFS 
GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/LyliaEngine/Pony_Diffusion_V6_XL

模型文件托管在 Hugging Face Hub，使用 HF CLI / SDK / Git 直接下载，不经过本站。

PyTorch / Transformers 使用

安装 Transformers

安装 Transformers 
pip install -U transformers torch

模型加载和推理

模型加载和推理 
from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained('LyliaEngine/Pony_Diffusion_V6_XL')
tokenizer = AutoTokenizer.from_pretrained('LyliaEngine/Pony_Diffusion_V6_XL')

完整文档

来源: HuggingFace

---
tags:

  • text-to-image

  • stable-diffusion

  • lora

  • diffusers

  • template:sd-lora

widget:
  • text: >-

score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up,
source_furry, beautiful female anthro shark portrait, dramatic lighting,
dark background
output:
url: images/00056-4212032008 (1).jpeg
base_model: Bakanayatsu/Pony-Diffusion-V6-XL-for-Anime
instance_prompt: None
license: cdla-permissive-2.0
---

Pony_Diffusion_V6_XL

<Gallery />

Model description

!tmpwcaemsmp.png
!ponydiffwall2.png

Pony Diffusion V6 is a versatile SDXL finetune capable of producing stunning SFW and NSFW visuals of various anthro, feral, or humanoids species and their interactions based on simple natural language prompts.

CHECK &quot;ABOUT THIS VERSION&quot; ON THE RIGHT IF YOU ARE NOT ON &quot;V6&quot; FOR IMPORTANT INFORMATION.

Please join our Discord Server to support development of new versions of this model and get access to free SD bot and check out more examples of this model capabilities on our prompt sharing website or follow the author on Twitter.
Important information

Make sure you load this model with clip skip 2 (or -2 in some software), otherwise you will be getting low quality blobs.

This model supports a wide array of styles and aesthetics but provides an opinionated default prompt template that allows generation of high quality samples with no negative prompt and otherwise default settings

score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up, just describe what you want, tag1, tag2

(previous Pony Diffusion models used a simpler score_9 quality modifier, the longer version of V6 XL version is a training issue that was too late to correct during training, you can still use score_9 but it has a much weaker effect compared to full string. You can learn more about these tags here).

The model is designed to not need negative prompts in most cases and does not need other quality modifiers like &quot;hd&quot;, &quot;masterpiece&quot;, etc...

Other special data selection tags include, &#39;source_pony&#39;, &#39;source_furry&#39;, &#39;source_cartoon&#39; and &#39;source_anime&#39; and ratings of &#39;rating_safe&#39;, &#39;rating_questionable&#39; and &#39;rating_explicit&#39;.

This model is capable of recognizing many popular and obscure characters and series.

If you are looking specifically for pony style, I recommend using one of the two following templates &#x60;anthro&#x2F;feral pony, rest of the prompt&#x60; or &#x60;source_pony, rest of the prompt&#x60;.

This model is trained on combination of natural language prompts and tags and is capable of understanding both, so describing intended result using normal language works in most cases, although you can add some tags after the main prompt to boost them.

Using Euler a with 25 steps and resolution of 1024px is recommended although model generally can do most supported SDXL resolution.

This model will sometimes generate pseudo signatures that are hard to remove even with negative prompts, this is unfortunately a training issue that would be corrected in future models. If that&#39;s an issue for you I suggest trying V5.5 or inpainting.

Special thanks

Iceman for helping to procure necessary training resources

Haru for assistance with captioning efforts

Cookie for technical expertise in training

PSAI Server Subscribers for supporting the project costs

PSAI Server Moderators for being vigilant and managing the community

Technical details

The model has been trained on ~2.6M images aesthetically ranked based on authors personal preferences, with roughly 1:1 ratio between anime&#x2F;cartoon&#x2F;furry&#x2F;pony datasets and 1:1 ratio between safe&#x2F;questionable&#x2F;explicit ratings. About 50% of all images has been captioned with high quality detailed captions, which results in very strong natural language capabilities.

All images has been trained with both captions (when available) and tags, artists&#39; names have been removed and source data has been filtered based on our Opt-in&#x2F;Opt-out program. Any explicit content involving underage characters has been filtered out.
License

This model is licensed under a modified Fair AI Public License 1.0-SD (https:&#x2F;&#x2F;freedevproject.org&#x2F;faipl-1.0-sd&#x2F;) license.

The following modifications have been added to Fair AI Public License:

You are not permitted to run inference of this model on websites or applications allowing any form of monetization (paid inference, faster tiers, etc.). This applies to any derivative models or model merges.

If you want to use this model commercially, please reach us at [email protected].

Explicit permission for commercial inference has been granted to CivitAi and Hugging Face.

Source

https://civitai.com/models/257749

Credit

https://civitai.com/user/PurpleSmartAI

Trigger words

You should use None to trigger the image generation.

Download model

Weights for this model are available in Safetensors format.

Download them in the Files & versions tab.

查看官方来源