Tang Yuan (汤圆Mix) – v2.0

Tang Yuan (汤圆Mix) - v2.0

Run Model 使用GRAVITI Diffus 关于GRAVITI Diffus
Model Name Tang Yuan (汤圆Mix)
Model Type Checkpoint
Base Model SD 1.5
Version Name v2.0
File Name tangYuanMix_v20.safetensors
SHA256 d3f1735c2b
NSFW False
Trigger Words
Tags anime, scenery, illustration, artstyle, background, wonton-based, backgrounds
Metadata fp = fp16
size = pruned
format = SafeTensor
Size 3.95 GB
Version Description

Added some LoRAs. Mostly background related + detailedeyes

Source https://civitai.com/models/47464
Author Jemnite


I tried out Dark Sushi 大颗寿司Mix (aka Dark Souls mix because it’s pronounced dàkēshòusī in Chinese lol) and thought it was fantastic. That’s based on Five Nuts Mooncake 五仁月饼Mix which is in turn a mix of Three Delicacy Wonton (三餡馄饨Mix) (Obligatory self-plug), ReV Animated, and 啥玩意完犊子(大概是一种古早画风)- old fish (I don’t understand this name tbh). That gave me an idea.

I have a model based on ReV Animated. Why not try mixing it with Wonton? Hence, 汤圆Mix. SomethingV3B sweet and sticky yet wet and soupy at the same time. 水水汤汤甜甜黏黏,这就是汤圆!

Huggingface repository


As far comparisons to Wonton go, TangyuanMix is SanXianWonton = lines + color – watercolor. If you just want Wonton with brighter more vivid colors, Ashen’s ColorBox color-enhancement of WontonV1 is still the way to go, but if you’re looking for something that captures more of WontonV2’s soft linework while still adding color (just less vivid) this is a good middle ground.


Use it however you would like. It is a generalized model. I recommend trying to prompt for beautiful backgrounds with it, since it has a lot of silicon and dpep4 lineage, it should be good at scenery. Theoretically.

@ayyyy22002 我终于用了这个汤圆的名字。你满意吗?


© 版权声明