Sign In

Toka paint style AnimaB1

5

Not recently loaded

Download

1 variant available

bf16 SafeTensor

Toka_paint_style_AnimaB1-03.safetensors

BF16, good balance • 3.34 MB

Verified:

Type
LyCORIS
Stats

54

5

3

Reviews
Published

Oct 29, 2026

Base Model
Training
Steps: 3,735Epochs: 3
Hash
AutoV228C1DA1E8A
Trigger Words
@toka paint

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

07363-1120557915-masterpiece, best quality, _lora_Toka_paint_style_AnimaB1-03_1_ @toka paint,_1girl, mita _(miside_), crazy mita _(miside_),_medi.png

The style of Toka paint.

Trigger word is @toka paint. polka dot background consistently evokes the author's characteristic diagonal dot pattern. Works best at lower resolutions (<1024), since most of the author's works are such. Trained on all the author's comics and commissions published on Telegram. The dataset was fully SFW.

The dataset was very big this time, so even just three epochs resulted in almost 4000 steps. I wanted this to also function as a charpack for Toka's OCs (I did tag them in the dataset), but they don't seem to have been memorized well. I've also tried doubling dim and alpha for this LoKr, but the size barely changed. I'm yet to figure out how to make a heavy LoKr and how good it would be at learning things.

If you like the style, please support the original author