2026/09/11 (040-041)
回归舒适区！还是单张的画的爽，这次尝试了下4096x4096，我和电脑都燃尽了，如果喜欢记得看原图，其他图也是，网站默认的是重新采样的图，和原图看起来还是不一样的。

Back to my comfort zone! Single images are still way more fun to make. Tried 4096×4096 this time, and both my PC and I are completely burnt out lol. If you like it, make sure to check out the original image! Same goes for my other works. The site displays a resampled version by default, which can look different from the original.

多来点r18g）

Base +6

被触手玩弄的是who，李小狼么

Base +4

这脚和手很细致，特别是原图，构图和背景也不错，现在AI已经这水平了吗？不过这惨目忍睹的分数…AI外加题材小众没办法，加油，画？的不错

Base +11

我超威，一边看着女友被ntr，一边被女友拳交，不敢想有多舒服

Base +1, magicat +7

Two or four images are not the best choice, having looked at other galleries, I can say that those that provide 40-60 images per mini-story are very popular.

Base +6

@GDRNET 谢谢鼓励！其实我生成每一张图片都很耗时间的。大体上有三步，第一步是实现初步大方向的构图和选角度，这一步要不停改提示词不停生成许多备选图片，然后根据接下来想要补充的细节和讲述的故事选出一张，第二步做画质提升的同时增加细节，第三步做局部重绘进一步增加细节，这一步我会特别细致的修改手脚。所以你确实可以说我在“画”图吧，哈哈哈。至于分数，不必在意，有人喜欢自然就有人不喜欢。

@VikAZ Thanks for the advice! Actually, my generation workflow is quite time-consuming because I put a lot of effort into maintaining good anatomy and details. So generating 40 images for a single story would take a huge amount of time unless I lowered the quality of each image, which I don't want to do. For now, I'll focus more on creating single images (or a few images at a time) and trying to tell a story through the details. I've actually already been trying to tell a whole story through things like expressions, backgrounds, and other small details, but I guess I'm not very good at it yet, haha. At my current stage, I'd like to experiment with different compositions and angles to get more practice. In the future, though, I may try creating longer stories with larger sets of images.

Base +1, strus +16, magicat +7

这个真不错吧！

Base +5

what prompts did you use?

Base +5