Sign In

Seraziel Style for Anima

1

Updated: Nov 15, 2026

styleserazielanima

Recently loaded

Download

1 variant available

bf16 SafeTensor

Seraziel Lora for Anima_merged.safetensors

BF16, good balance • 132.24 MB

Verified:

Type
LoRA
Stats

44

1

Reviews
Published

Nov 15, 2026

Base Model
Training
Steps: 4,750Epochs: 12
Usage Tips
Strength: 1
Hash
AutoV23682D320E4
Trigger Words
velqoraseraziel

The CircleStone Model is licensed by CircleStone Labs LLC under the CircleStone Non-Commercial License. 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

ComfyUI_temp_sbmkk_00009_.png

It seems that no one has made a Seraziel LoRA for Anima yet, so I've taken that challenge upon myself.

The model is a merge between 4250, 4500 and 4750 step saves. Use the base anima model for this one to work as you see on the examples, because it does not seem it pairs well with others without losing the style.

Steps, sampler, scheduler - to be frank, I found that to almost not matter with Anima so use whichever things you like. I personally have settings like these:
Sampler: euler a
Scheduler: simple
Steps: 25
CFG: 5

Trigger words for this model are velqora and seraziel.
It's not ideal. It's not overcooked, but I find the style to get lost when the LoRA is used with characters that are often depicted in a specific style. I wasn't able to reproduce the brushy strokes, messy lines. The light is a bit exaggerated. And a fair warning, Seraziel draws only with simple backgrounds most of the time, so model tends to... just not draw whatever background you want it to show. But still, I think its a good result.

Either way, let me know what you think, and hit me up if you want to commision a LoRA by any chance.