You can use the online workflow below. If the results are good, you can then try setting it up and running it locally.
👉 Click Here For Online WorkFlow
As a fan benefit, you’ll receive 1,000 points upon registration, plus 100 points for daily logins. Enjoy a smooth experience with an RTX 4090 and 48GB of VRAM.
Model Description
This is a LoRA specialized in Asian female style, trained for Wan2.2 T2V.
It works well for both text-to-image and text-to-video tasks, producing natural and refined results.
Recommended Settings
High-noise version weight: 1.2
Low-noise version weight: 1.2
Use Cases
✅ T2I generation
✅ T2V generation
✅ Combine with other LoRAs for unique styles
模型介绍
这是一个专注于 亚洲女性风格 的 LoRA，针对于 Wan2.2 T2V 进行训练。
无论是文生图还是文生视频，都能呈现自然、精致的效果。
推荐设置
高噪版权重: 1.2
低噪版权重: 1.2
使用场景
✅ 文本生成图片 (T2I)
✅ 文本生成视频 (T2V)
✅ 与其他 LoRA 搭配实现个性化风格探索
Description
FAQ
Comments (13)
牛逼
what the difference in high and less noise?
You use them in conjunction with each other. I think every single lora maker needs to include a workflow for Wan2.2 in their profile and model details page. WAN2.2 has confused a lot of people and it is a bit cumbersome to use. It's not very flexible when using multiple loras either. I've noticed that quality drop significantly even if you put the 2nd and 3rd lora to 0.25 strength it'll cause artifacts.
Wan2.2 uses MoE (Mixture of Experts) structure with a high noise 14B model and a low noise 14B model. The high noise model emphasizes global structure and motion while the low noise model emphasizes detail restoration and fidelity. Or think high noise as sketching, low noise as polishing. To train a LoRA for wan2.2, you need to train both the high noise and the low noise models to achieve the best result.
DaddyWolfgang aldniki217 I am not using COmfyui. I am using wan on pinokio. So should I use low to make it easier. I am new at 2.2.
AIBeauty4Seen when you use pinokio, you use Comfyui, it's just an other way to install. Use both, High and Low for better result, some workflow and loras are good with only Low, but try :)
AIBeauty4Seen The low-noise model of WAN2.2 should be very similar to WAN2.1. I haven’t tried using a LoRA for 2.2 on the 2.1 model, but it should work fine. You can give it a try.
aldniki217 porkepik863 Thank you both! :D
AIBeauty4Seen you are welcome :)
aldniki217 Hello, similar problem here: I'm using the online service "Wave Speed" with generation panel WAN 2.2 with LORA. It works giving the URL of the LORA and i've already tested it with a single LORA with success. User can also link more URL to add various LORA. But now, how it should work with HIGH and LOW LORAs? I can only add both but there is no way to specify "where". (is not like inside ComfUI)
settima_ai I haven’t used WaveSpeed myself, so I can’t provide a detailed answer. However, if one of the models you’re using works (it should be the low noise LoRA, right?), then just stick with that one.
对她没感觉
"Girl Like CDL"
May I ask what is the meaning of "CDL" ?
I dont' get it...
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.