I just made a new Telegram Group and Discord group drop in for chats, questions, feedbacks or just to share your work. 欢迎加入纸飞机和Discord群组。

RH体验地址（邀请码Invite Code: cptpovgo）：https://www.runninghub.ai/user-center/1993809672580788226?inviteCode=cptpovgo
喜欢我的作品吗？🎨 我会持续免费分享更多优质内容✨，欢迎支持一下，请我喝杯咖啡吧！💖 Like my work? 🎨 I share all my work for free ✨, so support me by buying me a coffee 💖! Wanna leave me a tip? Drop by my discord channel for links and special perks. 想赞助一下？来我 Discord，里面有链接和特别福利哦！

CivitAI 邀请码 Referral Code: R44VYP99 Extra 2.5k Buzz on membership purchase，买会员送2千5蓝电!

Ko-fi: ko-fi.com/catlover1937

Please do not redistribute without consent. 如非作者同意禁止搬运到其他平台，如国内Up主需要转发网盘请备注出处。

📌 TL;DR: Use V7.0 or the Official versions for general-purpose use. 日常使用推荐 V7.0 或 其他正式版本。

🎨 V7SE and the other Special Editions are mainly fun showcases of my recent work and experiments. They’re usually heavily modified in terms of styles, characters, and special concepts.

✨ They’re fun to experiment with, but generally not recommended as everyday, general-purpose models

🎨 V7SE 以及其他 Special Edition 主要是用来展示我近期的一些新作品和实验成果。它们通常会针对风格、角色以及特殊概念进行较大幅度的修改。

✨ 这些版本很适合拿来尝试一些有趣的新效果，但一般来说，并不适合作为日常的通用模型。

分享个我最近发现的app: ComfyLink,用手机远程跑自己 ComfyUI 的 App,还带有不错的提示词管理器。iOS 安卓都有,免费上手:[https://comfylink.app/?utm_source=civitai_catlover1937]

Sharing an app I've been really into lately — ComfyLink. It lets you run your own ComfyUI right from your phone, and it comes with a genuinely nice prompt manager. iOS & Android, free to start: [https://comfylink.app/?utm_source=civitai_catlover1937]

2026-08-27

🔥 Moody Krea2 Mix — Premium Edition

Hello there! 👋

It’s been a pretty fucking crazy few months dealing with model theft and all the bullshit that comes with it. 💀

But honestly, I want to start by giving a huge THANK YOU to everyone who helped me investigate this stuff. ❤️

With the help of people doing comparisons, technical analysis, and model-weight comparison exercises with me, we’ve managed to take down 15+ stolen / unauthorized models across multiple platforms, including a couple of unauthorized merges on Civitai. 🔬

And ironically… some of those stolen models had more favorites and usage than my actual official models. 😂

Funny enough, right?

Even worse, some people were actually making money by claiming my work as their own.

I’ve been releasing models for free for years, never really intending to make serious money from this hobby. So seeing people take that work, re-upload it, claim it as their own, and profit from it has honestly been pretty fucking disappointing.

You’d probably be shocked to know that there are also reputable creators on CivitAI — people you definitely know — who have been stealing my work. Honestly, that’s one of the most disheartening parts of all this.

But anyway…

Enough about that. Back to the real business. 😎

🔥 Introducing: Moody Krea2 Mix — Premium Edition

I’m finally bringing a new exclusive Premium Edition of Moody Krea2 Mix to my supporters.

I’ve actually been working on this since May, and holy shit… the amount of work turned out to be way more than I expected. 😂

As a one-man army, there’s only so much time I have in a day.

For this edition, I built a completely new NSFW / body anatomy dataset from scratch.

The dataset was constructed from 300+ realistic photosets, containing thousands of high-resolution images.

And no, this wasn't just a matter of throwing everything into a folder and hitting Train. 😂

Every single image was:

🖼️ Hand-picked for quality, composition, and diversity
🧹 Carefully processed with identity information removed
🏷️ Manually reviewed and retagged to improve prompt adherence
🔍 Filtered and refined multiple times throughout training
🧬 Selected to provide much broader face, body, and anatomy diversity

The final dataset contains 100+ different faces, a wide range of body shapes and proportions, and thousands of carefully curated images.

And in total, this model has gone through 90,000+ training steps.

Yes…

90,000+. 💀

And that number includes several failed experiments along the way.

There were runs that simply didn't work. There were datasets that had to be replaced. There were images that needed to be removed. There were tags that had to be completely reworked.

So I ended up retraining, testing, rebuilding, retagging, and retraining again… and again… and again. 🫠

At this point, I haven't done any gaming since May. 😂🎮

But after all that pain, I finally reached a result that I’m genuinely happy with.

And you guys know me — I’m ridiculously picky about my own models.

I’ve tested this thing from pretty much every angle I could think of, and the result has finally reached the level I was hoping for. 🔥

I really hope you guys enjoy it as much as I do.

❤️ A Thank You to My Supporters

More than anything, I want this release to be my way of showing my sincere appreciation to everyone who has supported me over the past couple of years since I came back to making models.

Seriously.

I wouldn't have gotten this far without you guys. ❤️

Whether you've downloaded my models, tested them, shared generations, given feedback, helped with troubleshooting, or simply stuck around and watched this whole thing evolve — it all means a lot to me.

This Premium Edition is something I wanted to make specifically for the people who have continued to support my work.

Private download links will be posted shortly on Discord. 📥

🔍 About Model Theft

And as always, if you come across something that looks suspiciously similar to my work, please let me know.

Don't assume something is mine or someone else's just from the filename or description.

Send it to me and I can investigate further.

There are actually quite a few ways to identify model lineage — both technically and aesthetically — and after doing all these comparisons recently, I've gotten pretty good at spotting them. 🔬

The Moody style is also… let's just say not exactly subtle. 😂

So if you see something suspicious, feel free to send it my way.

💎 About the Premium Edition

This particular model will remain pay-to-download.

I know this is a change from how I've released most of my models in the past, but after everything that's happened recently, I think it's the right decision for this particular release.

That said, I don't want to abandon free releases.

I promised you guys that I would keep releasing free models, and I intend to keep that promise. ❤️

But only if I would to continue on this journey, or not...

Anyway, that's enough talking.

Thank you again for all the support, patience, testing, and everything else you guys have done for me over the years.

Now go make some cool shit. 😎🔥

Have fun generating, and love ya. ❤️

🔥 Moody Krea2 Mix - 尊享版 Premium Edition

大家好！👋

最近这段时间真的有点疯狂。

一直在处理模型被盗、被搬运、被冒名顶替，还有各种乱七八糟的破事。💀

不过首先，我还是想认真感谢一下这段时间帮我调查这些事情的人。❤️

在大家帮忙进行模型对比、技术分析，以及各种 Model Weight Comparison 之后，我们目前已经成功让 15+ 个盗版 / 未授权模型在不同平台下架，其中还包括 Civitai 上几个未经授权的 Merge 模型。🔬

而最讽刺的是……

其中有几个盗版模型的 收藏数和使用量甚至比我自己的官方模型还高。 😂

是不是挺搞笑的？

更离谱的是，有些人甚至直接把我的模型说成是自己的，然后拿去赚钱。

我从重新开始做模型以来，基本一直都是免费发布，从来没想过靠这个兴趣爱好赚多少钱。

所以看到自己花了大量时间做出来的东西，被别人直接拿走、冒充原创，然后拿去盈利……

说实话，真的挺让人心凉的。

你可能很难相信，CivitAI 上甚至有一些你绝对认识的知名创作者，也在盗用我的作品。说实话，这才是整件事情里最让我心寒的部分之一。

我的好友 @Loraholic 最近也因为模型盗用的问题，停止公开分享他那些优秀的作品了——说真的，这对整个社区来说都是一个巨大的损失，因为那是一个真正不可替代的创作者。

老实说，我现在也开始认真考虑，自己是不是还应该继续公开分享作品。这样的事情真的非常令人失望和沮丧。最初我只是希望能够出于善意，把自己的作品分享给社区，但现在这一切已经完全偏离了我当初分享作品的初衷。

不过好了。

这些破事先放一边。😎

回到真正重要的事情。

🔥 Moody Krea2 Mix — 臻享版

今天，我终于可以把一个准备了很久的东西带给一直支持我的朋友们：

Moody Krea2 Mix — 臻享版

其实这个版本从 五月份就已经开始准备了。

本来以为只是做一个新的版本，结果最后发现……

妈的，这工作量完全超出了一个人的承受范围。😂

毕竟我是一个人搞这些东西，从数据集、训练、测试、调整，到最后的发布，基本全部自己来。

所以这次我是真的花了非常非常多的时间。

🧬 全新的 NSFW / Body Anatomy Dataset

这一次，我从零开始重新构建了一套全新的 NSFW / Body Anatomy 数据集。

整个数据集来自 300+ 个高质量真实 Photoset，最终整理出了 数千张 HD 图片。

而且不是简单地全部丢进去训练。

每一张图都是我自己筛选的。

我做了：

🖼️ 逐张人工筛选，确保画质和内容质量
🧹 处理并移除身份识别相关信息
🏷️ 人工审核和重新检查 Tags
🔍 多轮筛选和清理
🧬 增加不同脸型、身材、体型和身体比例的多样性
🎯 针对 Prompt Adherence 对标签进行了大量调整

最终的数据集里包含了 100+ 不同面孔，以及非常广泛的 不同身材、体型和身体比例。

而整个模型累计训练量超过：

90,000+ Steps 💀

没错。

九万多步。

当然，这里面也包括了一些失败的训练。

有些 Run 的结果不对，我直接废掉。

有些 Dataset 不够好，我重新换。

有些图片需要删除。

有些 Tags 需要重新整理。

然后重新训练。

再测试。

再修改。

再重新训练。

……如此循环。🫠

所以如果你发现我这几个月几乎没怎么出现……

因为我从五月开始基本就没玩过游戏了。😂🎮

🔥 但最后，结果终于让我满意了

经过这么多轮测试之后，我终于得到了一个 达到自己预期标准的结果。

而你们也知道……

我对自己的模型真的非常挑剔。

很多时候你们觉得已经很好了，我自己还是会觉得：

“嗯……还不够。”

😂

所以这一次，我是真的反复测试了非常久。

目前的最终结果，终于达到了我自己比较满意的程度。

希望你们也会喜欢。❤️

❤️ 给一直支持我的你们

其实这次发布，对我来说不只是一个新的模型版本。

更多的是想借这个机会，认真感谢一下过去这几年一直支持我的人。

从我重新开始做模型以来，已经过去了挺长一段时间。

一路上有人下载我的模型、帮我测试、分享作品、提供反馈、帮我解决问题，也有人一直默默关注着我做的东西。

如果没有你们，我绝对走不到今天。 ❤️

所以这个 Premium / 臻享版，也是我想给一直支持我的朋友们的一点回馈。

Private Download Links 会很快发布到 Discord。 📥

🔍 关于模型盗用

另外，还是和以前一样。

如果你们之后发现任何一个模型，看起来和我的作品非常可疑地相似，欢迎直接发给我。

不用自己判断是不是我的。

你把模型丢给我，我来查。 🔬

经过最近这段时间做的大量模型对比，我现在已经积累了一套比较成熟的判断方法。

很多时候，通过 Weight、结构、训练痕迹，以及最终生成风格，其实可以找到非常明显的证据。

而且 Moody 的风格本身就比较独特……

所以有时候真的很好认。😂

💎 关于这个版本

这个版本会保持 Pay-to-Download。

我知道这和我过去一直免费发布模型的方式不太一样。

但经历了最近这些事情之后，我觉得至少这个版本，我想把它作为一个 真正属于支持者的 Exclusive Release。

不过放心。

如果我决定继续开发下去的话，我会继续免费发布模型。

我之前答应过大家，以后还会继续做免费的版本。

这个承诺不会变。❤️

但是，能不能坚持下去却是另一回事了。

好了。

这次就不继续废话了。

真的非常感谢过去几年一直支持我的所有人。

感谢你们陪我一路做到现在。

也感谢那些愿意帮我一起抓盗版、做测试、分析模型的人。❤️

现在……

去玩模型吧。😎🔥

祝大家生成开心。

Love ya. ❤️

2026-08-18

Moody Krea2 Mix V7 Special Edition｜反差萌妹 2.0 发布啦！💗

当初发布反差萌妹 1.0 的时候，其实就收到了很多群友的赞赏，也一直有人私信问我会不会放出第二版。

现在，它来啦！🎉

这是基于 V7 的基础上做的萌妹化加强版本，并重新扩展了数据集。还是用代号的形式简单说一下：

Project MM（萌妹 LoRA）：数据集从 300 张提升到了 600 张，并重新进行高精度打标和训练。

Project Moody Aesthetic：这是 Moody 风格的主要核心，也是权重最大的一个 LoRA。真实感和细致感基本就靠它了。这次我把数据集从 300 张提升到了 1000 张，大幅度加强了光影、采光、构图、服装等方面的多样性。

最新的 Project Moody Aesthetic LoRA 并没有在 V7 中使用，所以 V7 SE 的效果可以拭目以待哦 👀

Project TYJR：加入了我们的「反差萌妹体态 2.0」LoRA，也就是某些群友最喜欢的那种骚包小圆脸、脸型和体态。

我知道这个版本不会讨好所有人，但是喜欢的人应该会很喜欢哦。

比如（我）。😂

大家玩得开心！💗

Moody Krea2 Mix V7 Special Edition｜Reverse Cute Girl 2.0 is here! 💗

When I released Reverse Cute Girl 1.0, I received a lot of positive feedback from the community, and I've also been getting DMs asking if I would ever release a second version.

Well, here it is! 🎉

This is a more "cute girl" focused enhancement built on top of V7, with the dataset expanded and retrained. I'll keep using the project codenames to briefly explain the changes:

Project MM (Cute Girl LoRA): The dataset has been expanded from 300 images to 600 images, with new high-quality captioning and retraining.

Project Moody Aesthetic: This is one of the main cores of the Moody style and also the largest-weighted LoRA. A lot of the realism and fine details rely on it. This time, I expanded the dataset from 300 images to 1,000 images, with significantly more diversity in lighting, illumination, composition, clothing, etc.

The latest Project Moody Aesthetic LoRA was NOT used in V7, so I'm really looking forward to seeing what V7 SE can do. 👀

Project TYJR: Added our "Reverse Cute Girl Body Type 2.0" LoRA, featuring the face shape and body type that some of you seem to love the most — that cheeky, cute, round-faced look.

I know this version won't appeal to everyone, but I think the people who like it will REALLY like it.

For example... me. 😂

Have fun! 💗

2026-08-15
I keep hearing feedback that Krea2 Mix feels too AI-generated and too perfect. If you’re looking for something more raw, realistic, and nostalgic, maybe give my new Moody Amateur Mix a try. 📸
看了一下我的小黑子们的评论，都说我的Krea2 Mix 太 AI、太完美了。如果你想要一些更加原始、真实、带有怀旧感的照片质感，也许可以试试我全新的 Moody Amateur Mix。📸

2026-08-10

A quick note on model selection:

After extensive testing, I’ve found that there are still noticeable differences between the different model variants. In my opinion, purely from a quality perspective:

BF16 > FP8 > NVFP4 > INT8

ConvRot INT8 on Krea2 is known to perform poorly in some cases. If you’re interested in the technical details, you can look up Kijai’s findings on why this happens.


简单说一下模型版本的选择：

经过大量测试后，我发现不同量化版本之间依然存在比较明显的差异。单纯从画质表现来看，我个人的排序是：

BF16 > FP8 > NVFP4 > INT8

Krea2 的 ConvRot INT8 在一些情况下已知表现比较差。如果你对其中的技术细节感兴趣，可以去看看 Kijai 的相关测试和分析，里面有解释为什么会出现这种情况。

Moody Krea2 Mix V7 — Official Edition 🎉

Yes, I heard your feedback.

After looking through the feedback and doing a lot more testing, I realized that I had pushed too far when trying to tackle the skin wetness issue in the previous version, and the trade-off in overall image quality simply wasn't worth it.

So for V7, I decided to take a step back toward V4 and blend my wetness fix back in at a much lighter weight. The older V4 Loras were also replaced by the newly trained substitution.

The result has been quite appealing so far. I ran it through a 500-image test set, and I'm much happier with the overall balance between image quality, skin texture, detail, and wetness.

In short: less aggressive wetness correction, better overall image quality.
Sometimes taking one step back is the best way to move forward. 😸

⚠️ Known Caveats

There are still a few things to keep in mind:

1. NSFW influence is relatively strong

V7 has a stronger NSFW tendency than some of my previous versions. I'm already planning to release a less NSFW version of the same generation in the Cutie Mix V4, for those who prefer a more SFW-oriented experience.

2. Some wetness is coming back

This is unfortunately unavoidable with the current balance. By reducing the strength of the wetness fix, some of the original wetness behavior naturally comes back.

However, I believe this is a much better trade-off than sacrificing overall image quality just to eliminate it completely.

3. Skin can appear slightly more greasy

Compared with V6, which intentionally pushed toward very dry skin and highly detailed skin texture, V7 has a somewhat more natural/greasy skin appearance.

This is intentional to some extent, and is part of the trade-off I made for better overall image quality.

🔬 Behind the Scenes

1. The base training dataset has been significantly expanded

I've almost doubled the base training set compared with previous iterations.

The expanded dataset includes substantially more variety in:

👗 Clothing
🏙️ Scenes & environments
💡 Lighting conditions
📸 Photography styles
🧍 Poses & compositions
🌎 General visual diversity

The goal here was not simply to make the model "better" at one specific type of image, but to give it a much broader foundation to work with.

Final Thoughts

V7 is essentially my attempt to find a better middle ground.

V6 went very aggressively after the wetness problem, and while that solved part of the issue, it came with too much of a quality trade-off.

With V7, I decided to prioritize the overall image first, while still keeping the wetness improvements from my previous work in the mix — just at a much more conservative strength.

I'm quite happy with where this version landed after the 500-image evaluation, and I hope you will be too.

As always, you don't have to upgrade just because a new version exists. If an older version works better for your particular workflow or use case, there's absolutely nothing wrong with sticking with it. ❤️

Enjoy V7! 🐱

Moody Krea2 Mix V7 — Official Edition 🎉

是的，我听到了大家的反馈。

在看了大家的意见、并进行了更多测试之后，我也意识到，之前为了处理 皮肤湿润感（wetness）问题，我确实有点矫枉过正了。虽然湿润感的问题有所改善，但同时也牺牲了一部分整体画质，这个取舍其实并不太值得。

所以这次 V7，我决定退回到 V4 的思路，然后把我之前针对 wetness 做的修正，以一个非常轻的权重重新融合进去。V4融合的旧Lora也被新练的Lora替代掉了。

经过我这次 500 张图片测试集的测试，我对最终的平衡还是比较满意的。

简单来说就是：

少一点激进的 wetness 修正，换回更好的整体画质。

有时候，往后退一步，反而才能走得更远。😸

⚠️ 已知问题

当然，这个版本也不是完美的，目前还有一些需要注意的地方：

1. NSFW 倾向会比较强

V7 的 NSFW 倾向相比之前的一些版本会明显更强一些。

我计划在之后的 Cutie Mix V4 中发布同一代、但 NSFW 倾向更低的版本，如果你更喜欢偏 SFW、日常向的效果，可以使用 Cutie Mix V4。

2. 一些 wetness 会重新出现

这个基本是不可避免的。

因为这次我降低了 wetness 修正的强度，所以之前被压制掉的一部分湿润感自然会重新出现。

不过相比为了彻底消灭 wetness 而牺牲整体画质，我认为现在这个取舍要合理得多。

3. 皮肤会比 V6 更偏油润

和 V6 相比，V6 的皮肤明显更加干燥，并且非常强调细腻、丰富的皮肤纹理。

V7 则会稍微偏向一种更油润、更有光泽的皮肤表现。

这也是这次整体取舍的一部分。

🔬 幕后的一些变化

1. 基础训练集大幅扩充

这次我把基础训练集进行了大幅扩充，规模接近之前的两倍。

新增的数据包含了更多样化的：

👗 服装
🏙️ 场景与环境
💡 光照条件
📸 摄影风格
🧍 姿势与构图
🌎 整体视觉多样性

这次的目标并不是单纯针对某一个问题进行优化，而是希望给模型一个更加丰富、更加全面的基础。

❤️ 最后

V7 可以说是我这次尝试寻找的一个更好的平衡点。

V6 在处理 wetness 问题上走得比较激进。虽然确实解决了一部分问题，但同时也带来了比较明显的画质损失。

所以这次 V7，我决定把重点重新放回到整体画质上，同时保留之前针对 wetness 所做的一部分改进，只是把它的影响降低到更加保守的程度。

经过这次 500 张测试集的验证，我个人对 V7 最终的表现还是比较满意的。

希望大家也会喜欢这个版本！ ❤️

当然，和以前一样，新版本并不意味着旧版本就一定更好或者更差。如果你发现某个旧版本更适合自己的用途，那完全没有必要为了更新而更新。

找到最适合自己的版本才是最重要的。🐱

Enjoy V7! 🎉

2026-08-08

As I already mentioned the wetness lora a few times before in the previous release notes, I really want to make sure I give credit to @Loraholic again for his great work on the slider loras so his work doesn’t go unnoticed. I've integrated the wetness slider in my V6 with weight -1.5 and V7 with weight -0.3 (Upcoming) models.

正如我之前在几次 Release Note 中提到过的 Wetness LoRA，我还是想再次特别感谢 @Loraholic ，以及他在 Slider LoRA 方面所做的优秀工作。

在这里也再次明确注明一下：我在 V6 中使用了 Wetness Slider，权重为 -1.5；而即将发布的 V7 中则使用了 -0.3 的权重。

2026-08-04

我也不知道你们说V6的「毛刺感」到底是什么，是指这种吗？😂 如果是的话，这可是我花了很大功夫一点一点精调出来的效果啊……

For those saying V6 looks lower quality than V4, I'm genuinely confused. 😅 To me, this is exactly the look I've been trying to achieve after spending countless hours fine-tuning it.

皮肤细节 Skin Texture Demo (Euler Ancestral, bata, 8 Steps)

2026-08-03

⚠️ 关于 FP8 版本
Due to changes in my training pipeline, I will no longer be providing the official ComfyUI FP8 version.
Based on previous experience, converted FP8 models may occasionally suffer from image quality degradation or other unexpected issues. If you encounter any serious problems, please let me know so I can investigate.
In the meantime, I highly recommend giving the INT8 or NVFP4 versions a try—they've performed far better than I originally expected, both in quality and overall experience.
由于训练流程进行了调整，今后我将不再提供官方的 ComfyUI FP8 版本。
根据以往的经验，转换得到的 FP8 模型偶尔可能会出现画质下降或其他异常情况。如果你遇到比较严重的问题，欢迎及时向我反馈，我会继续调查和优化。
同时，我也非常推荐大家尝试 INT8 或 NVFP4 版本，实际效果真的比我预期的好很多，无论是画质还是整体体验都有不错的表现。

🚀 Moody Krea2 Mix V6 - Official Release 🌎✨

Hello everyone! 👋

After weeks of training, testing, and incorporating your feedback from the V6 Special Edition, Moody Krea2 Mix V6 is finally here! ❤️

First of all, I owe everyone an apology.

🙏 About the Wet Skin Fix

I need to correct something I mentioned in the V6 Special Edition (Asian Fever) release notes.

I previously said the wet skin issue had been fixed, but after reviewing everything, I discovered that I made a mistake during the model merging process, and the final V6 Special Edition unfortunately did not include the wet skin fix as intended.

Sorry for the false advertisement and for the confusion—that was entirely my mistake. 😅

🧴 This Time, Goodbye Wet Skin!

This is the version that finally includes the complete fix.

Over the past few weeks, I've spent a significant amount of time researching Krea2's notorious wet skin, random skin blemishes, and unwanted skin artifacts.

The skin rendering has been extensively retrained to produce smoother, cleaner, and more natural-looking skin while still preserving realistic texture and fine details.

Unless your prompt explicitly asks for it, the model should no longer randomly generate shiny, wet-looking skin or strange skin imperfections.

✨ Higher Quality Training

V6 is built using a significantly higher-quality training dataset than any previous release.

Combined with my newly retrained 1536×1536 foundation LoRAs, the overall image quality has taken a huge leap forward. You'll notice cleaner skin, richer textures, sharper facial features, improved clothing details, and much better overall realism.

🌍 Full Ethnicity Support

Unlike the Special Edition, which primarily focused on Asian aesthetics, the official V6 has been trained with full ethnicity support.

I also worked hard to keep the model as ethnicity-neutral as possible, allowing your prompts—not the model's bias—to determine the final appearance.

💎 Universal Beauty Enhancement

Instead of enhancing only one ethnicity, V6 introduces a universal beauty enhancement across all races while preserving each ethnicity's unique facial characteristics.

The goal is simple: make everyone look better—not the same.

🎨 Excellent LoRA Compatibility

Maintaining strong LoRA compatibility has always been one of my highest priorities.

V6 remains extremely flexible and works beautifully with a wide variety of style LoRAs, allowing them to express their intended style without constantly fighting against the base model.

⚙️ Recommended Settings

After extensive testing, my current recommended settings are:

Sampler: Euler A + BETA
Steps: 8

if you want more details, you can increase the steps up to 12, and also increase the SD upscale steps too.

This combination offers an excellent balance between image quality, prompt adherence, and generation speed, and has become my default recommendation for V6.

For the best overall experience, I also highly recommend using my 4K HD Workflow.

Since V6 preserves much more fine detail on its own, you can safely lower the SD Upscale Denoise value to around 0.15, allowing the workflow to retain significantly more of the original image while still producing crisp, natural high-resolution results.

Thank you to everyone who tested the V6 Special Edition and shared your feedback.

Every bug report, suggestion, and comparison helped shape this release into what I originally envisioned.

I sincerely hope you'll enjoy Moody Krea2 Mix V6.

Happy generating! ❤️

🚀 Moody Krea2 Mix V6 正式版 🌎✨

大家好！👋

经过这段时间不断训练、测试，以及吸收 V6 Special Edition 的反馈之后，Moody Krea2 Mix V6 正式版终于来了！ ❤️

首先，要先向大家道个歉。

🙏 关于 Wet Skin 修复

我要更正一下 V6 Special Edition（Asian Fever） 发布说明中的一个错误。

当时我提到已经修复了 Wet Skin 问题，但后来重新检查后才发现，我在模型合成（Merge）过程中出了纰漏，导致最终发布的 V6 Special Edition 并没有包含原本应该加入的 Wet Skin 修复。

很抱歉造成误导，也谢谢大家一直以来的反馈。这次确实是我的失误。😅

🧴 这一次，真的和 Wet Skin 说再见！

这次正式版才是真正包含完整修复的版本。

这段时间我花了大量时间研究 Krea2 一直以来令人头痛的 Wet Skin、随机皮肤瑕疵以及各种奇怪的皮肤细节问题。

我重新加强了皮肤相关训练，让皮肤更加细腻、自然，同时保留真实的纹理和细节表现。

默认情况下，如果提示词没有特别要求，模型将不会再随机生成湿润皮肤或各种奇怪的皮肤瑕疵。

✨ 更高质量训练

V6 使用了迄今为止质量最高的一套训练数据进行训练。

配合全新重训的 1536×1536 基础 LoRA，整体画质相比过去有了非常明显的提升，无论是皮肤细节、五官、服装纹理、光影层次还是整体质感，都更上一层楼。

🌍 Full Ethnicity 支持

不同于 Special Edition 主要针对亚洲审美，这次正式版加入了 Full Ethnicity（全人种） 训练。

同时我也尽可能降低模型本身的种族偏向，让不同族裔都能获得稳定、高质量的表现，把最终人物的外观更多交给 Prompt 来决定，而不是模型本身。

💎 全人种颜值增强

这次的人脸美化不再针对单一族裔，而是针对所有人种进行了整体优化。

我的目标不是让所有人长得一样，而是在保留各自种族特征的前提下，让每一种族都拥有更自然、更高颜值的表现。

🎨 更优秀的 LoRA 兼容性

LoRA 兼容性一直都是我最重视的部分之一。

V6 依旧保持了非常优秀的兼容性，各种摄影、动漫、写实以及不同风格 LoRA 都能够更加自然地发挥自身特色，而不会被基础模型过度干扰。

⚙️ 推荐设置

经过大量测试，目前我最推荐 V6 使用以下参数：

采样器： Euler A + BETA
步数： 8 步

如果想要更多细节，可以将步数增加到 12，也可以增加 SD Upscale步数。

这组参数在画质、提示词遵循度以及生成速度之间取得了非常好的平衡，也是我目前最推荐的默认配置。

另外，我也非常推荐搭配我的 4K HD Workflow 一起使用。

由于 V6 本身已经拥有更好的细节表现，因此 SD Upscale 的 Denoise 可以降低到 0.15 左右，这样能够保留更多原图细节，同时依然获得非常自然、干净的高分辨率效果。

最后，非常感谢所有参与 V6 Special Edition 测试并提供反馈的朋友。

每一条 Bug 回报、建议以及对比测试，都帮助我把 V6 打磨成了我最初想要的样子。

希望大家喜欢 Moody Krea2 Mix V6，祝大家玩得开心！❤️

2026-08-02

亚洲限定，

🔥 Moody Krea2 Mix V6 Special Edition: Asian Fever ✨

Hello everyone! 👋

Today I'm releasing something a little different—a Special Edition built alongside the development of Moody Krea2 Mix V6.

I actually debated whether this should simply become the Cutie Mix V4, but this model is much closer to the direction of the unreleased V6 (WIP). This versions specifically focused on Asian aesthetics, while the official V6 is still in training and will be a Full Ethnicity release.

I also wanted to use this opportunity to gather everyone's feedback and see if this direction is worth pushing even further.

✨ What's New?

🧴 Extreme enhancement on skin (Hopefully!)

One of my biggest goals over the past few weeks has been tackling Krea2's infamous wet skin and random skin blemish/wart issues.

To address this, I significantly strengthened the skin rendering during training, resulting in smoother, cleaner, and more detailed skin.

By default, unless your prompt specifically asks for it, the model should no longer introduce shiny skin or random skin imperfections.

📸 New Photography Style

This Special Edition also includes a brand-new photography-style LoRA that I've been training recently.

If enough people are interested, let me know in the comments—I may release it separately in the future!

😊 Enhanced Asian Beauty Model

I've further improved my Asian face enhancement model while also increasing facial diversity, helping characters stay beautiful without feeling overly repetitive.

🚀 Massive Quality Upgrade

This is probably the biggest quality investment I've made so far.

All three of my core foundation LoRAs have now been completely retrained at 1536 × 1536, bringing a substantial improvement to overall image quality, texture fidelity, and fine details.

I genuinely hope you'll enjoy the results.

If the community responds well to this release, there's a very good chance the official Moody Krea2 Mix V6 (Full Ethnicity Edition) will be arriving soon™.

As always, thank you for all the support, feedback, and bug reports—they genuinely help shape every new version. ❤️

🔥 Moody Krea2 Mix V6 特别版：Asian Fever ✨

大家好！👋

这次带来的是 Moody Krea2 Mix V6 Special Edition：Asian Fever！

其实我犹豫了很久，要不要直接把它当作Cutie V4发布。不过因为这个跟未发布的V6更加接近，所以我还是决定把它作为一个特别版独立放出。

它整体的训练思路已经非常接近正在训练中的 V6 正式版，但这次主要针对的是亚洲审美，因此没有加入多人种训练素材。正式版 V6 则会是 Full Ethnicity（全人种） 版本。

也希望借着这次发布，收集大家的意见，看看这个方向是不是大家喜欢的。

✨ 这次有哪些变化？

🧴 重点改善 Krea2 的皮肤问题

最近这段时间，我一直在研究如何解决 Krea2 那种恼人的皮肤湿润感（Wet Skin）以及各种随机出现的皮肤疣、皮肤瑕疵问题。

因此我专门加强了皮肤相关训练，让皮肤看起来更加细腻、光滑，同时保留更多真实纹理。

默认情况下，如果提示词没有特别要求，模型基本不会主动生成湿润皮肤或各种皮肤瑕疵。

📸 全新摄影风格

这次还加入了一个我最近新训练的摄影风格 LoRA。

如果大家感兴趣的话，欢迎留言告诉我，我会考虑后续单独发布出来。📷

😊 亚洲颜值进一步升级

我进一步强化了自己的亚洲人脸美化模型，同时尽量增加人物长相的多样性，希望在保持高颜值的同时，也不会让每个人都长得过于相似。

🚀 画质全面升级

这次是真的下了不少成本。😂

我把自己一直使用的三个基础 LoRA 全部重新训练到了 1536 × 1536 分辨率，相比过去，整体画质、细节表现、纹理层次都有了非常明显的提升。

希望大家能够喜欢这次的尝试！❤️

如果大家反馈不错的话，那么 Moody Krea2 Mix V6 正式版（Full Ethnicity Edition） 应该很快就会和大家见面！

感谢大家一直以来的支持，也欢迎继续给我各种反馈和建议，它们真的会影响后续版本的发展！🙌

2026-07-24

⚠️ Notice: If you've already downloaded the V5.0 Teaser, you don't need to download the final V5 release. After extensive testing, I decided to use the teaser as the official V5, as the higher-step training did not deliver the improvements I was hoping for.

⚠️ 提醒：如果你已经下载了 V5.0 Teaser（预览版），就无需再次下载 V5 了。经过进一步测试后，我决定直接将 Teaser 作为正式版 V5，因为更高步数的训练并没有带来更好的效果。

INT8 is CONVROT, I don't make regular int8 anymore. INT8就是Convrot，普通的int8我不再支持了。

✨ Moody Krea2 Mix V5 is Here! ✨

🚀 Major Update from V4

Hello everyone! 👋

Moody Krea2 Mix V5 is finally here, and this is one of the biggest updates since V4.

🌟 What's New

🖼️ Most of the training has been completely re-done at 1536px (up from 1024px, but yikes training is expensive now), allowing the model to learn much finer details and better overall aesthetics.
💖 Further enhanced visual aesthetics, producing cleaner, more polished, and more appealing images.
⚖️ NSFW LoRA influence has been slightly reduced. In V4 it was pushed a little too far, which occasionally affected overall image quality. V5 brings things back into better balance while maintaining excellent compatibility with NSFW LoRAs.

Overall, V5 should deliver more consistent quality while keeping the signature Moody look that many of you enjoy.

🔥 New workflow

🎉 A dedicated 4K SeedVR2 version has been released! Try it if you're looking for even cleaner high-resolution results.

As always, thank you everyone for your support, feedback, and testing. ❤️ Enjoy V5!

✨ Moody Krea2 Mix V5 发布啦！✨

🚀 V4 以来最大更新之一

大家好！👋

Moody Krea2 Mix V5 正式发布，这也是自 V4 以来最大的一次升级。

🌟 更新内容

🖼️ 大部分训练内容已重新制作，并提升至 1536 分辨率训练（V4 为 1024, 不得不说现在训练的成本真高），让模型能够学习更多细节，整体画质与质感进一步提升。
💖 美学表现进一步增强，整体画面更加精致、自然，也更耐看。
⚖️ 适当降低了 NSFW LoRA 的权重。 V4 的强度稍微有些过头，虽然效果明显，但也会对整体画质产生一些影响。V5 在保留兼容性的同时，取得了更好的平衡。

整体来说，V5 在保持 Moody 风格 的同时，也带来了更加稳定、更高质量的出图体验。

🔥 新工作流

🎉 4K SeedVR2 专用版本也已经发布！ 想要更高分辨率、更干净细节的朋友可以试试。

感谢大家一直以来的支持、反馈与测试！❤️ 希望大家喜欢 V5！

2026-07-21

I was planning to release a new 4K HD ultimate workflow to help reduce skin imperfections, but all of the demo images were created with the upcoming V5 model. Since I can't properly tag images with a model that hasn't been released yet, I decided to put out this V5 teaser instead.

I'm currently wrapping up V5 with a few final refinements. Once everything is polished and ready, I'll release it as a free download, just like always.

If you're a supporter, you can grab the direct download link from my Discord right away.

See you all soon for the official V5 release!

原本我是想发布一个新的4KHD 終極工作流，帮助减少皮肤瑕疵，不过演示图片全部都是使用即将发布的 V5 模型生成的。由于模型还没有正式发布，我没办法正确给这些图片打上模型标签，所以决定先放出这个 V5 抢先预览版（Teaser）。

目前 V5 已经进入最后的收尾阶段，我正在做一些最终的细节调整。等一切准备完成后，我会像往常一样免费开放下载。

支持者可以直接到我的 Discord 获取下载链接，抢先体验。

那么，我们很快就在 V5 正式版发布时见面！

2026-07-17

🌸✨ MOODY × CUTIE Edition is HERE! ✨🌸

📌 TL;DR

💙 Looking for a general-purpose Krea2 model?
➡️ Stick with V4.0.

💖 Love the Moody Cutie aesthetic and want enhanced NSFW performance?
➡️ This Special Edition is made for you.

👋 Hello again!

It's been a while since I released a Special Edition, and during that time I've received countless comments and DMs asking:

"Will there be another Special Edition?"

The answer is...

🎉 YES! 🎉

I know these Special Editions aren't as highly anticipated as the major version releases, but they've always had their own audience—and I truly appreciate everyone who's been waiting for them. ❤️

After all, beauty is subjective. Everyone has their own preferences, and different parts of the world have completely different standards of beauty. That's exactly why I enjoy making these niche editions.

💖 Introducing the MOODY × CUTIE Edition

Built on the solid foundation of Moody Krea2 Mix V4.0, this edition blends in several signature LoRAs from Moody Cutie Mix, while also featuring enhanced NSFW capabilities.

The goal isn't to replace V4.0.

Instead, I wanted to keep everything people loved about V4.0 —its quality, consistency, and versatility—while steering it closer toward the signature Moody Cutie aesthetic.

✨ Cuter characters
🌸 Softer moods
😊 Sweeter expressions
💖 More of that unmistakable Moody vibe

Hopefully I managed to nail the balance this time... 🤞

As always, thank you for all your support. I hope you enjoy this little passion project as much as I enjoyed making it.

Happy generating! ❤️

🌸✨ MOODY × CUTIE 特别版来啦！✨🌸

📌 一句话总结

💙 想要一款全能、通用的 Krea2 模型？
➡️ 请选择 V4.0。

💖 喜欢 Moody Cutie 的风格，并希望拥有更强的 NSFW 表现？
➡️ 那这个特别版就是为你准备的。

👋 好久不见！

距离上一次推出 Special Edition（特别版） 已经有一段时间了。

这段时间里，我收到了很多评论和私信，大家最常问的一句话就是：

"还会出特别版吗？"

答案当然是……

🎉 会！而且它来了！🎉

我知道，相较于每次的大版本更新，Special Edition 的关注度没有那么高。

但它一直都有属于自己的受众，而我也一直很感谢大家愿意耐心等待。❤️

毕竟，每个人对「美」的定义都不同，不同国家、不同文化，也都有自己喜欢的审美风格。

也正因为如此，我一直都很喜欢制作这种偏向个人风格的小众版本。

💖 MOODY × CUTIE Edition

本次特别版基于 Moody Krea2 Mix V4.0 打造，在保留 V4 优秀基础的同时，融合了 Moody Cutie Mix 中几款最具代表性的 LoRA，并进一步强化了 NSFW 表现。

它并不是为了取代 V4.0。

相反，我希望保留 V4.0 一直以来优秀的画质、稳定性与泛用性，同时让整体风格更加靠近我心目中的 Moody Cutie。

✨ 更可爱的角色
🌸 更柔和的氛围
😊 更甜美的表情
💖 更有辨识度的 Moody 风格

希望这次真的把这种感觉带到了 Krea2 上。🤞

最后，还是要感谢大家一直以来的支持。

希望你们会喜欢这次的特别版，也祝大家玩得开心！❤️

2026-07-16

Updated V4.0 Non-Comfy for people who are using WanGP/NEO and other tools that do not support FP8. 更新至 V4.0 非COMFY 版本，適用於使用 WanGP/NEO 和其他不支援 COMFY FP8 格式的工具的使用者。

2026-07-10

✨ V4.0 Model Update

A short update note but lots of small stuff! This version includes a newly trained LoRA along with many quality improvements.

🔹 What's Changed

🧠 Integrated a newly trained LoRA for better overall generation quality.
✨ Reduced unwanted skin texture imperfections while preserving natural-looking fine details.
🚿 Reduced the chance of random body fluid artifacts appearing in NSFW generations.
🎨 Improved compatibility and support for external LoRAs.
🚀 For the best experience, pair this model with the new Krea2 Simple Workflow V3 or Minimal Workflow V3. Together they deliver noticeably cleaner outputs, better clarity, and improved overall image quality.
Fixed the FP8 conversion quality degrade issue
Fixed blurry wormlike background issue

✨ V4.0 模型更新

短简介大更新！本次版本加入了全新训练的 LoRA，并带来多项画质优化。

🔹 更新内容

🧠 整合了全新训练的 LoRA，进一步提升整体生成质量。
✨ 在保留丰富细节的同时，减少皮肤纹理瑕疵，让肤质更加自然干净。
🚿 降低 NSFW 场景中身体随机出现体液残留伪影的概率。
🎨 提升对外部 LoRA 的兼容性与表现。
🚀 强烈建议搭配全新的 Krea2 Simple Workflow V3 或 Minimal Workflow V3 使用，可显著提升画面清晰度、干净度以及整体生成质量。
修复FP8转换降画质问题
修复背景失真问题

2026-07-07

🚀 I also added the INT8 CONVROT version! In theory, it should deliver quality similar to Q8 GGUF while running at speeds close to INT8. ⚡

🚀 我新增了 INT8 CONVROT 版本！理论上，它能够提供接近 Q8 GGUF 的画质，同时拥有接近 INT8 的运行速度。⚡

好像很多人不知道每个版本代表着什么，It seems like many of you are unsure of what the models variants mean so here's my take for it:

I was able to rent a cloud server and done a special merge on BF16. Supporters can download version BF16 of this model directly from my discord link. DM me for discord permission tags. 我成功租到了一台云端服务器，并完成了这个模型的 BF16 特别合并版本。支持者现在可以通过我的 Discord 链接直接下载 BF16 版本。如需获取 Discord 下载权限，请私信我领取对应的权限组（Tag）。

FAQ:

Where is int8?
The mxfp8 is the int8, because of civitAI doesn’t have an int8 model type in the dropdown.
What are the special editions (models without Version tag)?
TLDR: download official editions (ones with version tags) for overall better experience.
The Special Edition are heavily inspired by my Asian girl character and style LoRAs. It was never originally intended for release—I created it purely for own enjoyment. However, after sharing some sample results in the community, quite a few people asked if I could release it, so I decided to publish it for those who are interested. It’s designed for a niche audience that enjoys the 网感 (internet aesthetic) beauty filter look.

Hope that clears things up!

2026-07-06

TDLR: V3 is so powerful that i was impressed when testing it. You can safely delete the previous versions now.

✨ Moody Krea2 V3 Release ✨

Hey everyone! 👋

V3 is here, bringing another solid upgrade over V2 with a stronger focus on realism while keeping the signature Moody look. 📸

🔥 What's Improved

🧑‍🎨 Increased overall realism and more natural-looking results.
🌄 Fixed the lazy background rendering issue from V2. Many of you reported the "worm-like" texture that sometimes appeared in blurred backgrounds—this has been significantly reduced.
💃 Improved human anatomy and more natural poses.
📝 Retrained several LoRAs using both Chinese 🇨🇳 and English 🇬🇧 captions, giving the model a better understanding of prompts and noticeably improving prompt adherence.
✨ Enhanced skin texture for results that look less AI-generated and more photographic.

⚙️ Workflow Update

I've also made a few adjustments to my recommended workflow to better support the model and get the most out of V3.

👉 Please give the new workflows a try for the best experience!

As always, thank you all for the feedback and bug reports—they've been incredibly helpful in making this release possible. ❤️

Enjoy, and have fun generating! 🚀

✨ Moody Krea2 V3 发布啦！✨

大家好～👋

V3 是基于 V2 的全面升级版本，在保留 Moody 风格的同时，进一步提升了整体真实感。📸

我只能说V3。。。很强。。。

🔥 本次更新内容

🧑‍🎨 整体画面更加真实、自然。
🌄 修复了 V2 中较常见的背景偷懒渲染问题。不少用户反馈虚化背景偶尔会出现类似「蚯蚓纹」的纹理，现在已经大幅改善。
💃 优化人体结构与姿势生成，让人物更加自然。
📝 重新训练了多组 LoRA，并同时使用 中文 🇨🇳 + 英文 🇬🇧 Caption，模型对 Prompt 的理解能力更好，Prompt Adherence 也有明显提升。
✨ 优化皮肤质感，减少 AI 感，让成图更加接近真实摄影效果。

⚙️ 工作流更新

我也针对模型调整了推荐工作流，让 V3 能发挥更好的效果。

👉 建议大家使用新的 Workflow，以获得最佳生成体验！

感谢大家一直以来的反馈和 Bug 回报，很多改进都是来自社区的建议。❤️

祝大家玩得开心，Happy Generating！🚀

2026-07-04

🎉🌸 Special Edition Drop / 特别版发布 💖

🇺🇸 English

Hello everyone! 👋

Here comes another Special Edition. Say no more. 😏

This release is trained on a heavily Xiaohongshu-inspired dataset, featuring East Asian women with beauty-filter aesthetics and curvier proportions. It's made specifically for a niche audience that enjoys this particular style. 💕

If you're not the target audience, feel free to skip this one and wait for the official V3, which is already in the works and will be the more balanced release for everyone. 🚀

Cheers, have fun, and enjoy! 🍻✨

🇨🇳 中文

大家好！👋

又来一个 特别版 啦，不多说了。😏

这次依旧是一个 小红书风格浓厚 的数据集，主打东亚女生、滤镜感颜值，以及更加丰满的身材表现。💕 这是专门为喜欢这种风格的小众用户准备的版本。

如果这不是你的菜，也完全没关系～可以直接等待正在制作中的 官方 V3，那会是更加均衡、适合大众使用的正式版本。🚀

祝大家玩得开心，玩得尽兴！🍻✨

2026-07-01

🎉🚀 Moody Krea2 V2 Official Release! 🌍✨

The wait is over! 💙 Moody Krea2 V2 is officially here!

The Oriental Edition was just a special edition released exclusively for my lovely Discord & Telegram community ❤️. It was made for a specific audience.

The Global Edition should be considered the official V2, featuring the most balanced finetuning I've achieved so far while preserving the signature Moody quality. 😎✨

🎮 GL HF, and enjoy! ❤️

⚠️ PS: About the MXFP8 version

The MXFP8 upload is my attempt at providing an INT8 version. Since Civitai doesn't currently have an INT8 model option, I selected MXFP8 instead.

I followed all the available conversion steps, but I wasn't able to fully verify that it works in every environment. As a result, I can't guarantee compatibility or functionality, so please use it at your own discretion.

🙏🎉🚀 Moody Krea2 V2 正式发布！🌍✨

终于来了！💙 Moody Krea2 V2 正式版 现已发布！

之前的 萌妹版版（Oriental Edition） 只是专门为 Discord 和 Telegram 群友准备的特别版本 ❤️，面向的是比较特定的受众。

而这次的 Global Edition 才是 真正的 V2 正式版 🌎，也是目前为止我调校得最均衡、最能体现 Moody 风格 的版本。😎✨

🎮 祝大家玩得开心，Enjoy！ ❤️

⚠️ PS：关于 MXFP8 版本

MXFP8 是我尝试制作的 INT8 版本。由于 Civitai 目前没有 INT8 的模型选项，所以只能选择 MXFP8 作为上传类型。

我已经按照目前已知的流程完成了转换，但 没有办法完整验证它是否能在所有环境下正常运行。因此，我无法保证兼容性或功能正常，如果有问题还请自行承担风险，感谢理解！🙏

2026-06-30

🌸 Krea2 V2 – Oriental Edition is Here! | Krea2 V2 东方版正式发布！ 🎉

🇺🇸 English

✨ Krea2 V2: Oriental Edition is here!

This SPECIAL EDITION was actually requested by members of our chat group. I originally made it just for my own entertainment and wasn't planning to release it, but after seeing how many people wanted it, I decided to share it with everyone. ❤️

This version focuses on beautiful East Asian female aesthetics with a strong Instagram and Xiaohongshu (RED) inspired vibe. 📸🌸

🔥 What's New

🤝 Huge kudos to @salttaro ! His Better Pussy & Poses V4 plays a major role in this edition on NSFW.
🎨 Added 2 newly trained Moody Aesthetic LoRAs, ported from ZImage for richer visuals.
🔞 Enhanced NSFW prompt adherence for better consistency and prompt accuracy.
NVFP4 Version.

🚀 A regular Krea2 V2 release is coming soon, so stay tuned!

Thank you everyone for your support! ❤️

🇨🇳 中文

✨ Krea2 V2 东方版 正式发布！

其实，这个版本是应聊天群里不少朋友的要求制作的。原本只是做给自己娱乐使用，并没有打算公开发布，不过看到大家都很期待，所以决定分享给大家。❤️

本版本主打 东亚女性美学，整体风格拥有浓厚的 Instagram 与 小红书（Xiaohongshu） 氛围，适合喜欢清新、时尚、高质感人像的用户。📸🌸

🔥 更新内容

🤝 特别感谢芋头哥 @salttaro ！本模型融合了 Better Pussy & Poses V4，对整体效果帮助非常大。
🎨 加入了 2 个全新训练的 Moody Aesthetic LoRA，并从 ZImage 移植整合，让画面氛围更加出色。
🔞 增强 NSFW Prompt 遵循能力，提高生成稳定性与提示词响应效果。
NVFP4 格式

🚀 普通版 Krea2 V2 也将在近期发布，敬请期待！

感谢大家一直以来的支持！❤️

2026-06-25

YES - IT SUPPORTS ALL RACES - I AM JUST LAZY BUILDING A MULTICULTURAL PROMPT DB.

配套工作流 Paired Workflow.

RH体验地址：https://www.runninghub.ai/zh-cn/post/2070402594421436417/?inviteCode=cptpovgo

🚀 Moody Krea 2 Mix V1 is HERE! 🎉🔥

After lots of testing, it's finally time to release the very first version of Moody Krea 2 Mix! 💫

This is the initial release—I'm testing the waters with Krea 2, but I have to say... this base model is absolutely amazing. 🤯✨

I've already got several improvements and future training plans in the works, so this is just the beginning! Expect even better quality and refinements in upcoming versions. 💪🎨

🔓 NSFW is fully unlocked.

Have fun, experiment, and let me know what you create! Your feedback will help shape future updates. ❤️

Enjoy! 🚀🔥

🚀 Moody Krea 2 Mix V1 正式发布！ 🎉🔥

经过一段时间的测试，终于迎来了 Moody Krea 2 Mix 的首个版本！✨

这是一次初始版本，先放出来试试水。不过不得不说，Krea 2 的底模真的非常惊艳，潜力十足！🤯

后续我已经规划好了几轮新的训练和优化，所以这只是个开始，未来还会继续提升画质、细节和整体表现，敬请期待！💪🎨

🔓 NSFW 已完全解锁。

欢迎大家尽情体验、多多出图，也欢迎分享你的作品和反馈，你们的建议都会成为后续更新的重要参考！❤️

祝大家玩得开心！🚀🔥

PS: Workflow in this image - 图里有工作流 I will be posting an official workflow shortly. 我晚点发配套工作流。

Files with SHA256: 12d42c1f43d94bbbee1cc1ffaf8df1ad31a8cbeb9da7cd90e44f06405feaa1ca