Sign In

Hela : Death’s Elegant Daughter

383

Download

2 variants available

Type
LoRA
Stats

106

0

Reviews
Published

Jul 9, 2026

Base Model

Anima

Hash
AutoV2
8B0A6DE2E1
Trigger Words
Hasti
default creator card background decoration
Followers - 2082

2.1K

Likes - 4515

4.5K

License:

Anima

The Anima Model is licensed by CircleStone Labs LLC. Copyright CircleStone Labs LLC. IN NO EVENT SHALL CIRCLESTONE LABS LLC BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL.

Built on NVIDIA Cosmos

Anima_12436_.png

Cold beauty, sharp eyes, and a presence that feels supernatural . works better for gothic portraits .


Illustrious Lite Version update note :

After an unreasonable amount of testing, sacrifices, and probably a few lost brain cells, this version is ridiculously more adaptable to different checkpoints and LoRAs. It plays nice with almost anything you throw at it without suddenly forgetting its style or insisting on doing the same three boring poses over and over again.

You can even crank weights past 1.0 like a reckless scientist and, surprisingly, the results usually are sane.

Of course, no upgrade comes without a patch note labeled "known issues."

The trade-off is that this version is a little less obsessed with preserving every microscopic facial characteristic. You may occasionally notice slight variations in facial features, and sometimes her body type decides to exercise its creative freedom by changing ever so slightly.

Fortunately, there's an easy fix. In my testing, adding the following tags DRAMATICALLY improves consistency and helps her remember who she is:

aged-up, milf, almond eyes, big eyes, dark green eyes, brown hair, thick bold eyebrows, arched brows, full lips, defined cupid's bow, straight nose, high cheekbones, sharp jawline

Think of them as gentle reminders rather than strict instructions. Like showing an actor their own headshot before they walk on stage.


Please support the model with a like, reaction, or comment. Training these things isn't exactly cheap, and every interaction helps fund future experiments, poor financial decisions, and occasionally even new models.