v2.0将基础模型切换为我的新模型“Veriwarp”。此版本需要完全不同的使用方式。所以请务必阅读模型说明和生成设置，或者直接使用例图中的工作流。

v2.0 switches the base model to my new model, "Veriwarp." This version requires a completely different approach, so please be sure to read the model description and generation settings, or use the workflow from the sample images directly.

与v1.3相比

优点：更灵活的年龄tag，更好的提示词遵从度，更明显的画风还原度。

代价：偶尔更差的肢体

Compared to v1.3

Pros: More flexible age tags, better prompt adherence, and more distinct art style reproduction.

Trade-off: Occasionally poorer anatomy.

目前支持的年龄tag

女性：lxlx、mesugaki、bishoujo、lady、mature lady

男性：shxtx、bishounen、gentleman

请您用以上tag代替1girl、1boy。

(请注意年龄tag仍不稳定，角色年龄可能会随着提示词的变化而改变，如果发生影响可以加重年龄tag的权重，已对此进行优化)

Currently Supported Age Tags

For females: lxlx, mesugaki, bishoujo, lady, mature lady

For males: shxtx, bishounen, gentleman

Please use these tags instead of 1girl or 1boy.

(Note: The age tags are still unstable, and the character's age may change with different prompts. If this becomes an issue, you can increase the weight of the age tag. Optimizations have been made to address this.)

正面提示词中不要填写任何质量tag，例如“masterpiece、best quality、highres、absurdres、newest、 very aesthetic、4k、8k、HDR”等等都不需要。

负面提示词:"masterpiece,best quality,worst quality,low quality,watermark"。

是的，您没有看错，这并不是写错了，还请在负面提示词中填写以上tag。

关于外貌穿着类提示词：请尽量不要同时使用含义重复的的tag。例如dress、long dress、blue dress、frilled dress，请直接写成blue frilled long dress。

Positive Prompts

Do not use any quality tags in the positive prompt, such as "masterpiece, best quality, highres, absurdres, newest, very aesthetic, 4k, 8k, HDR," etc. None of these are needed.

Negative Prompts

"masterpiece, best quality, worst quality, low quality, watermark"

Yes, you are reading this correctly. This is not a mistake. Please include the tags above in your negative prompt.

Regarding Prompts for Appearance and Clothing

Please avoid using tags with overlapping meanings. For example, instead of dress, long dress, blue dress, frilled dress, combine them directly into blue frilled long dress.

生图分辨率

分辨率应介于1024*1024（总像素1,048,576）-1536*1536（总像素2,359,296）之间的任意尺寸。

低于1024*1024（总像素1,048,576）通常会使图像变差，高于1536*1536（总像素2,359,296）不会让图像产生质变。

个人经常使用960*1344/1096*1536。还请您根据图片复杂度进行调整，并不是越大越好。

Generation Resolution

The resolution should be any size between 1024x1024 (1,048,576 total pixels) and 1536x1536 (2,359,296 total pixels).

Resolutions below 1024x1024 generally result in a poorer image, while resolutions above 1536x1536 will not yield a qualitative improvement.

I often use 960x1344 or 1096x1536. Please adjust according to the image complexity, as bigger is not always better.

采样器与调度器

er sde/kl optimal，如果您不清楚这是什么还请进行尝试！

CFG：4~6都可以正常工作。

步数：30~40步是一个通常的设置。

Sampler and Scheduler

er sde / kl optimal is recommended, if you are unfamiliar with these, please experiment with them!

CFG Scale

A range of 4 to 6 works well.

Steps

A typical setting is 30-40 steps.

为了应对civitai和tensor日益严格的审查，为了模型能够长久存在。 现在决定将Mature ritual修改为MRitual。 感谢大家一直以来的支持。 也希望大家在使用在线生成时尽量不要生成NSFW内容，以避免模型被后台系统标记为风险。 In response to the increasingly strict censorship on Civitai and Tensor, and to ensure the model's long-term survival. I have decided to rename "Mature ritual" to "MRitual". Also you please refrain from generating NSFW content when using online generation features, to prevent the model from being flagged as a risk by the backend system. Thank you all for your continued support.