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。
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 💖!
Ko-fi: ko-fi.com/catlover1937
Fanvue: https://www.fanvue.com/catlover1937
Please do not redistribute without consent. 如非作者同意禁止搬运。
2026-05-20
Hey lads & gents 🍻
V7.0 just dropped 🚀
✨ Minor optimizations for the newer Moody Releases models
🛠️ Added automatic detailer module
🎭 Added manual masked detailer module
🌊 Re-organized the workflow layout to feel more like a waterfall flow
⚡ Plus a bunch of QoL improvements inspired by my latest workflows
Enjoy & cheers ❤️
🇨🇳
兄弟们来了 🍻
V7.0 现已发布 🚀
✨ 针对新版 Moody Releases 模型进行了小幅优化
🛠️ 新增自动细节修复模块
🎭 新增手动蒙版细节修复模块
🌊 重新整理了工作流布局，让整体流程更像瀑布式结构
⚡ 同时加入了不少我最新工作流里的 QoL 改进
玩得开心，干杯 ❤️
2026-03-31
V6.0 ZIT Simple Dual Ksampler Enhanced WITH SDA workflow.
DEMO images were generated by moody porn v10 model.
预览图生成于 moody porn v10 model.
With the help of SDA, the new workflow works almost as well as my ZIB+ZIT workflow just with minor drawbacks of bad poses and body anatomy. This is still a very good solution for the “Diversity Collapse” problem.
Z-Image-Turbo-SDA is a highly efficient LoKr (Low-Rank Kronecker Product) adapter designed to rescue the “Diversity Collapse” problem in few-step distilled Flow Matching / Diffusion models. However, it also increases the chance of bad body anatomy.
SDA Lora support added in 1ST sampler
New upscaler used for further attempt to eliminate the side effect of SD upscale
QoL updates as my other newer workflows
在 SDA 的帮助下，这个新的工作流表现几乎和我的 ZIB + ZIT 工作流一样好，只是存在一些小缺点，比如姿势不自然和人体结构不准确。不过，这仍然是解决“多样性崩塌（Diversity Collapse）”问题的一个非常有效的方案。
用于解决少步数蒸馏 Flow Matching / Diffusion 模型中的“多样性崩塌”问题的高效 LoKr 适配器，但也会在一定程度上增加人体结构异常的概率。
增加了对SDA LORA 的支持
使用了新的放大器（SD upscaler），进一步尝试消除 SD 放大带来的副作用 （莫名的水渍，鸡皮疙瘩，油光）
加入了一些使用体验上的优化（QoL），整体流程更加顺手，并与我其他较新的工作流保持一致
2026-03-20
V5.0 ZIT Dual Ksampler Enhanced Simple workflow
V5.0 is here. Some optimization and QoL update for the Moody ZIT and ZID models better support. Added an attempt to fix some anomolies from SD upscale node.
V5.0 版本已发布。针对最新的moody模型 进行了一些优化和体验改进。尝试修复SD upscale奇怪纹路的问题。
2026-02-25
Dual K-Sampler setup with minor quality of life updates with support of my ZIT/ZID (Zimage distilled) models.
Availble ZIT Models / 可选 Turbo 模型：
https://civarchive.com/models/620406/moody-porn-mix
https://civarchive.com/models/621441/moody-real-mix
https://civarchive.com/models/2384856/moody-wild-mix?modelVersionId=2681752
2026-01-20
Add separate seeds for each Ksampler so you don't have to re-run the whole flow from beginning if you just want minor changes
Use a different upscalers for better performance and quality
Included the facedetailer flow by default
Added optional SeedVR for upscaling to 4k
Added optional skin contrast filter
2026-01-13
Updated with 2 Ksamplers for better preview and optimization.
2026-01-01
Added a face detailer workflow as requested. Since some users have noted that character LoRAs are not fully compatible, my recommended workaround is to first describe the character using a text prompt, then apply the face detailer with your character LoRA to effectively perform a “face swap.”
--
As requested, here is a simplified, minimal workflow derived from my setup for the most basic Z-image AIGC tasks.
Description
FAQ
Comments (31)
I'm using the workflow in question and generally getting great results, but a message about missing Lora in the code lines caught my attention.
‘’[rgthree-comfy][Power Lora Loader] Lora “zimage\characters\Pico-v4.safetensors” not found, skipping.
[rgthree-comfy][Power Lora Loader] Lora “zimage\characters\ZIT66.safetensors” not found, skipping."
Are these Lora files necessary? I searched for their names online but couldn't find them. If they are needed, is there a link where I can find them?
Or can I continue without them?
No, those are just my character loras I trained for personal use, and I put them in for example, slot in your own if needed.
Thank you for this amazing workflow.
Just want to ask, where can I download the Diffusion Model: zimage\moody-V9.safetensors.
Thanks a lot.
Sorry that is actaully something I am in work with. Just use the latest model from Moody porn mix or Moody real mix
I have this error with V3:
SimpleMath+module 'ast' has no attribute 'Num'something wrong with the simple math module it seems, try updating it, if still no luck, you can just remove them
请问放大模型哪里下载阿？
求图生图版本的
把empty latent image改成 load image，denoise设置0.6 就行。
加载放大模型早哪里下载
This workflow is a memory pig!!! Tried my best to reduce it's consumption but oh wow
move the seedvr into a separate workflow maybe?
If you're using ComfyUI look into the MultiGPU nodes by pollockjj, there's a small fix in the issues pages #167. It will let you create virtual VRAM from system RAM instead of doing offloads and partial loads for loading bigger models
这个工作流是不是得32g显存才行？我这16g直接慢的离谱。
哪一步慢？ 16 够够的
It is generally running great. For the Face detailer, I'm getting the below error for the node. Would you know how to fix it?
UltralyticsDetectorProvider
Weights only load failed. This file can still be loaded, to do so you have two options, [1mdo those steps only if you trust the source of the checkpoint[0m. (1) In PyTorch 2.6, we changed the default value of the weights_only argument in torch.load from False to True. Re-running torch.load with weights_only set to False will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source. (2) Alternatively, to load with weights_only=True please check the recommended steps in the following error message. WeightsUnpickler error: Unsupported global: GLOBAL ultralytics.nn.tasks.DetectionModel was not an allowed global by default. Please use torch.serialization.add_safe_globals([ultralytics.nn.tasks.DetectionModel]) or the torch.serialization.safe_globals([ultralytics.nn.tasks.DetectionModel]) context manager to allowlist this global if you trust this class/function. Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html.
are your models .pt
Not to my knowledge. I'm mainly just using your workflow and your v8 checkpoint though I'm running it on a ZImage turbo runpod template
@fancypants2789872 you need specific fave detection model in ultralytics folder
慎下，seedvr2会导致你加载时间非常离谱的长，一模一样的工作流，5090dv2。64g内存，最新版comfyui
你第一次用没模型他要下载肯定长啊呢 🤦
@catlover1937 写教程还得写保姆级，把用户都当xx才行，哈哈哈
@qoob9006 就差没喂饭了
seedvr2放大的时间不是特别长，但是你选的模型初次运行没有下载它会下很长时间，如果你的梯子和网速不行，下载到本地了就快了。
光有配置不会玩怎么行。
请教一下，这个提示词是怎么写的，要用什么格式给打标的软件，我打出来都是这样：
图中是一个东亚女性，身着白色短袖衬衫，搭配灰黑相间的格纹领结，下装为黑灰格纹百褶裙。她留着深棕色长发，发间系着黑色大蝴蝶结。她双手轻托一枝开满白色樱花的花枝，花枝上花朵繁茂，花瓣轻盈，部分花苞含苞待放。背景中可见模糊的自然景致，远处似有水面与绿意，营造出清新柔和的春日氛围，整体画面传递出静谧美好的感觉。
额，一两句话说不完，可以加群问