Type | |
Stats | 3,777 |
Reviews | (478) |
Published | Jun 28, 2023 |
Base Model | |
Hash | AutoV2 7B4F7CFF33 |
求求了!来点返图啊啊啊啊啊啊啊啊啊啊
来点评论啊啊啊啊啊啊啊
这一版的画风相比T-comic更平面,更日漫化,而且细腻的多!算是一次不错的尝试吧!和美漫特化的T-comic可以区分开了。
The style of this version is more flat, more diffuse, and much more detailed than the T-comic! It's a good try! It can be distinguished from the T-comic of the COMIC。
v2这个版本强化了画场景的能力(至少不会一团糟),然后增强了图像的稳定性(v1在某些长tag下容易糊)
V2 This version enhances the ability to draw scenes (at least not a mess), and then enhances the stability of the image (v1 tends to paste under some long tags)
v3这个版本强化了语义识别敏感度,加强了肢体结构的线条,强化了脸部的特效以及眼睛的精细程度,而且最最最关键的是!我剪枝了,使得模型变得非常小。(如果模型有什么问题一定一定要在评论区说)
This version of v3 strengthens the sensitivity of semantic recognition, strengthens the lines of the limb structure, enhances the special effects of the face and the fineness of the eyes, and most importantly, the most important! I pruned so that the model became very small. (If there is any problem with the model, be sure to say it in the comment area)
v4这个版本吸收了一些T-comic的数据,然后改善了一些v3里很恶心的肢体问题,剪枝后出现的问题还是蛮多的,感觉v3的使用不如v2,所以推出了v4,出图更加有逻辑
This version absorbed some T-comic data, and then improved some of the disgusting limb problems in v3, and there were quite a lot of problems after pruning, and I felt that the use of v3 was not as good as v2, so v4 was launched, and the plot was more logical