Type | |
Stats | 1,876 |
Reviews | (245) |
Published | Dec 19, 2023 |
Base Model | |
Hash | AutoV2 9DFAB17ADF |
🖥️Welcome to try out the open-source GPT4V-Image-Captioner, developed by my friend and me. It offers a one-click installation and comes integrated with multiple features including image pre-compression, image tagging, and tag statistics. Recently, we also launched the webui plugin version of this tool, everyone is welcome to use it!
🌍欢迎加入QQ群"兔狲·AIGC梦工北厂"，群号 ：780132897 ；"兔狲·AIGC梦工南厂"，群号 ：835297318（入群答案：兔狲）。Telegram群聊“兔狲的SDXL百老汇”，链接：https://t.me/+KkflmfLTAdwzMzI1
Welcome everyone to the fantasy world of AIArt! This is my second SDXL base model work after HelloWorld SDXL Realistic base model. This model is trained based on a variety of different styles of illustrations/anime works that I have been accumulating for a long time. It is compatible with a variety of flat minimalist styles, and its tonal style stands out among the mainstream 2D/2.5D large models that have been released.
Everyone is welcome to experience and use it, and provide as much feedback as possible. Your suggestions are very important for improving the model in the next step!
Recommended parameters for using the model:
Steps ≥ 25
CFG scale: 7
Size ≥ 1024x1024
If you encounter a very satisfactory image, it is recommended to use Hires. fix to redraw at a 0.3-fold magnitude and enlarge by 1.5 times. The recommended enlargement algorithm is: R-ESRGAN 4x+ Anime6B.
欢迎大家来到AIArt的幻想世界！这是我继HelloWorld SDXL写实大模型后的第二个SDXL大模型作品。本模型基于我长期积攒精选的多种不同画风插画/二次元作品训练而来，兼容多种扁平简约画风，调性风格在已发布的主流2D/2.5D大模型中独树一帜。
欢迎大家体验使用，多多返图反馈。大家的建议对模型下一步的改进非常重要！
模型建议的使用参数：
Steps ≥ 25
CFG scale: 7
Size ≥ 1024x1024
如果roll到很满意的图像，建议使用Hires. fix以0.3倍重绘幅度进行1.5倍放大，建议的放大算法：R-ESRGAN 4x+ Anime6B