AiARTiST Art Composition 艺术海报构图设计模型 SD V3版本 LoRA

SD v3 Poster Composition Design LoRA

Due to the good response to the previously released version 1.5 "AiARTiST Art Poster Composition Design Model",After half a year, we finally ushered in the "fragmented" SD3 m2b open source version. In order for the community to continue the carnival,By the way, I reunited the broken mirror, so I started to participate in the fine-tuning competition of SD3. The current version is my first SD3.

LoRA is used to test the training effect of SD3 Diffusers, and it is hoped that the community will make progress together.

Since the community released the v3 weight on June 12, we have preemptively released the workflow and model files.

SD3-CADS-Workflow https://www.liblib.art/modelinfo/09ca04ae8dfe41b887fdaf1580bd93a4

Currently, it supports Wenshengtu, Tushengtu, Wensheng Video and v3 LoRA loading. Although we are not the most perfect,But we will definitely try our best to be as down-to-earth as possible so that everyone is not limited by equipment, video memory, and computing power.

Experience the feeling of running the latest v3 large model smoothly locally as soon as possible.

Subsequently, while posting a large number of SD3 example images, we also released XL LoRA that simulates SD3 image quality and style.

AiARTiST-SD3 Simulator-XL https://www.liblib.art/modelinfo/7449642bdf39443aa66341837025c690

It is used to compare and contrast the image quality differences between XL and SDv3m. On the one hand, it can accelerate everyone's understanding of the new features of the SD3 version.

On the other hand, we also look forward to the continued growth of the XL version’s vitality.

AiARTiST 艺术海报构图设计模型 ArtComposition SD V3版本 LoRA

SD v3海报构图设计LoRA是什么？

由于之前发布的1.5版本《AiARTiST 艺术海报构图设计大模型》反响不错，

时隔半年，我们终于迎来了“支离破碎”的SD3 m2b开源版，为了社区能继续狂欢，

顺便让破镜重圆，所以开始参与SD3的微调竞赛，目前这个版本是我的第一个SD3

LoRA，用于测试SD3 Diffusers的训练效果，抛砖引玉，希望社区一同进步。

自6月12日，社区释放v3权重以来，我们抢先发布了工作流和模型文件。

SD3-CADS-工作流  https://www.liblib.art/modelinfo/09ca04ae8dfe41b887fdaf1580bd93a4

目前已经支持 文生图、图生图、文生视频 以及 v3 LoRA加载。虽然我们不是最完善的，

但是我们一定会尽量做到最接地气，让大家尽量不用受设备、显存、算力的限制，

尽快体验到最新的v3 大模型在本地畅快的跑起来的感觉。

随后，我们在大量PO出SD3例图的同时，又发布了XL模拟SD3画质和风格的LoRA

AiARTiST-SD3模拟器-XL https://www.liblib.art/modelinfo/7449642bdf39443aa66341837025c690

用来对比和关差XL和SDv3m 之间的画质差异，一方面加速大家对SD3版本新特性的认知，

另一方面，也期待着XL版本的生命力能够持续高涨。
----------------------------------------------------------------------

目前这个LoRA还只是一个测试版，画面质量和泛化性能都非常有限！

Rank 128 / lr 0.0001 / 跑了1万步（发布的是6000步版本）

推荐底模：sd3_medium_incl_clips 5.6GB版本（无T5）

推荐权重：0.6~0.8  CFG：3~4.5

推荐采样器：DPM++2m 调度器：sgm_uniform

推荐步数：28~30

推荐工作流：SD3-CADS-LoRA工作流 上面有链接

推荐尺寸：768 x 1536 / 640 x 1280 / 576 x 1024
----------------------------------------------------------------------

目前这个LoRA还只是一个测试版，画面质量和泛化性能都非常有限！

Rank 128 / lr 0.0001 / 跑了1万步（发布的是6000步版本）

推荐底模：sd3_medium_incl_clips 5.6GB版本（无T5）

推荐权重：0.6~0.8  CFG：3~4.5

推荐采样器：DPM++2m 调度器：sgm_uniform

推荐步数：28~30

推荐工作流：SD3-CADS-LoRA工作流 上面有链接

推荐尺寸：768 x 1536 / 640 x 1280 / 576 x 1024

Model list: https://www.liblib.art/search?keyword=AiARTiST

Free models account for 55% and open source models account for 95%

Member models account for 40%, closed source models account for 5%

Commercial model:

AiARTiST-CADS2.0 XL commercial advertising auxiliary system (enterprise customization)

https://www.liblib.art/modelinfo/a56ebacdba7d4e30b97bb124bc3fc28f <<Click here -------------------------------------------------- ------------------- Model purpose statement: 1. You may not host this model and its derivatives (such as the Fusion Model version) on websites/applications that plan to earn revenue or donate. 2. You are not allowed to directly sell this model and its derivative versions (such as the fusion model version) unless you have made sufficient manual modifications to the model so that it can be fully recognized as your personal work in a legal sense. If you violate this article, you will be personally responsible for all legal consequences, and I will not be held responsible. 3. You cannot use this model to intentionally create or share illegal or harmful content and output. Please abide by public order and morality and use this model for positive purposes. -------------------------------------------------- ------------------- Please leave a message if you have any test questions. For business cooperation, please see your personal homepage +V Zyuan980 AiARTiST | Metaverse China·Shandong Digital Man/AiGC/Metaverse Direction Be a good tool person and serve artists. Contact information: https://afdian.net/a/AiARTiST-lib

----------------------------------------------------------------------

模型用途声明：

1. 您不得将此模型及其衍生版本（如融合模型版本）托管于计划赚取收入或捐赠的网站/应用程序。

2. 您不得直接售卖此模型及其衍生版本（如融合模型版本），除非您对此模型进行了足够程度的人工修改，使其在法律意义上可以被完全判定为您的个人作品。如果您违反本条，所造成的一切法律后果由您个人承担，请恕本人概不负责。

3. 您不能使用该模型故意制作或共享非法或有害的内容传播和输出，请您遵守公序良德，将此模型用于积极正面的用途。

----------------------------------------------------------------------

测试问题请留言，业务合作看个人首页 +V Zyuan980

AiARTiST | Metaverse 中国·山东 数字人/AiGC/元宇宙方向

做好工具人 服务艺术家 联系方式：https://afdian.net/a/AiARTiST-lib