CivArchive
    Preview 3615077
    Preview 3001293
    Preview 3615075
    Preview 3615076
    Preview 3615078
    Preview 3000368
    Preview 3000542
    Preview 3000541
    Preview 3000577
    Preview 3000578
    Preview 3000615
    Preview 3000624
    Preview 3000744
    Preview 3000748
    Preview 3000750
    Preview 3000911
    Preview 3000913
    Preview 3000944
    Preview 3000965
    Preview 3000967

    QQ交流群753289747,群内会不时发布内测模型

    我的游戏已上线Steam商店页面,求求各位大佬加个愿望单~↓↓↓↓↓
    My game has been launched on Steam! Wishlist it now if you're interested! ❤️↓↓↓↓↓

    Steam 上的 真相剪辑师(Redacted)

    【强烈推荐】最新版2D Pixel Toolkit (2D像素工具包)

    九个不同的Lora,角色和序列帧动画,冷兵器,枪械,植物,动物,书籍,药水瓶和等距建筑

    Nine distinct Lora, character and frame-by-frame animation, cold weapons, firearms, plants, animals, books, potion bottles, and isometric architecture.

    专门为独游开发训练的模型,花费了半个多月的心血,原本是打算自用不公开的,但后来想到独游事业的发展,还是公开吧,希望能够帮助到大家,游戏上线了评论里说下我可能会去玩哈哈

    直出的像素图是真·像素图,不是对不齐格子的伪·像素图,并且边缘清晰,不会模糊,也没有污染,对其他Lora的兼容性也都OK,可同时使用其他人物Lora,强烈推荐配合插件palettize使用,可以直出像素图↓↓↓

    GitHub - Astropulse/sd-palettize:Automatic1111 扩展,用于对生成的图像进行调色

    大模型推荐 anything4.5anylora,还有ReV Animated

    如果只是想出像素图,关键词是 pixel,pixel art,pixelart,xiangsu,xiang su

    如果想出横版2D的像素人物,请务必在上面的关键词后面再加上 ((Side view)),full body,solid background

    Originally, I intended to use it for myself and not make it public, but later I realized that the development of the indie game industry needs it, so I made it public, hoping it can be helpful to you.The game has been launched and the comments say that I may go to play haha.

    The pixel image produced is a true pixel image, not a pseudo-pixel image that is not aligned with the grid. The edges are clear, not blurry, and there is no contamination. It is also compatible with other Lora models and can be used simultaneously with other character Loras. 

    GitHub - Astropulse/sd-palettize:Automatic1111 扩展,用于对生成的图像进行调色

    Model recommendation anything4.5,anylora or ReV Animated
    If you only want to create pixel images, the keywords are 
    pixel, pixel art, pixelart, xiangsu, xiang su
    If you want to create a side view, full body, solid background pixel character, be sure to add 
    ((Side view)), full body, solid background after the above keywords.

    Description

    FAQ

    Comments (13)

    TijuanaSlumlordOct 17, 2023· 2 reactions
    CivitAI

    Thx, works with all LORAs and models. Bit of work in Photoshop and good to go.

    huzhehuanOct 18, 2023
    CivitAI

    你是对的,好用的工具确实能帮到很多独游开发者,支持你的免费分享👍

    2027324Oct 18, 2023
    CivitAI

    666,下载了,向大佬膜拜

    wyxzddsjj919Oct 18, 2023
    CivitAI

    虽然不知道大佬在说什么,但是对于喜欢合金弹头的我来说,给你点赞,谢谢分享

    (黑尼玛性感!)

    400010Oct 18, 2023
    CivitAI

    支持!感谢你的分享!

    a2890655840859Oct 21, 2023
    CivitAI

    大佬我的超人!!

    zz101ks888Nov 1, 2023
    CivitAI

    巨赞,正需要这个,大佬请收下我的祝福

    JellaiNov 3, 2023
    CivitAI

    The results are really good! I'm striving for grid accuracy in my own model, but am having a bit of trouble. How did you train this? Were your training images all the same original resolution? Did you resize smaller images to larger before training? Or did you train smaller images? Thanks!

    SYK006
    Author
    Nov 4, 2023· 1 reaction

    Hello, I am not very good at English, so my answer is translated using ChatGPT.

    All my training images are grids 64*64 pixels, and they are the same size, 515*512. After training, I also utilized lora hierarchical technique for optimization.Hierarchical optimization is a key link, which is related to the compatibility of lora. It took me a long time to complete this lora..

    sansmiaNov 7, 2023· 6 reactions
    CivitAI

    Wow. I am absolutely blown away by the quality this produces. I have tried so many pixel Loras in the past and while some of them work pretty well, all of them have had caveats I had to work around. With this one, I can keep bumping the resolution and it just creates higher resolution pixel art without distortion. Incredible work.

    SYK006
    Author
    Nov 8, 2023

    Thank you for your review ; )

    cyusyn2Nov 8, 2023
    CivitAI

    模型的 Output name好像有问题,输出的是:<lora:pixel_f2_0.5_lbw=ONLY+OUTS+像素小人Pixel Character_1 0.3+pixel-portrait-v10.75:1> 而且这样不能触发

    SYK006
    Author
    Nov 8, 2023

    模型直接使用是pixel(11)这个名字

    LORA
    SD 1.5

    Details

    Downloads
    31,678
    Platform
    CivitAI
    Platform Status
    Available
    Created
    10/17/2023
    Updated
    6/19/2026
    Deleted
    -
    Trigger Words:
    pixel,pixel art,pixelart,xiangsu,xiang su