Sign In

Eira Kishida 2: Rakuen Shinshoku: Island of the Dead (Illustrious/Anima)

5.2k

Early Access

Loaded

Download

1 variant available

bf16 SafeTensor

Eira_Kishida_Verion_1_Anima_anima_3395385_epoch_12.safetensors

BF16, good balance • 132.24 MB

Verified:

Type
LoRA
Stats

1

28

Generation

Loaded

Reviews
Published

Oct 9, 2026

Base Model

Anima

Training
Steps: 1,200
Epochs: 15
Hash
AutoV2
8ED3F5AF59
Trigger Words
eira_kishida
rakuen shinshoku: island of the dead
natural beauty, natural breasts, dark brown hair, short hair, bangs, green eyes, dark brown eyebrows, perfect eyebrows, perfect face, perfect eyes, light brown skin, big breasts,
anime coloring, anime screencap
default creator card background decoration
Followers - 1152

1.2K

Likes - 1463

1.5K

Downloads - 10049

10K

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

generator_import_1791591640728_1.jpg

More words have been added to the training data and the colors should now be a bit more vibrant.

If you like the character, always leave a like and a picture, and if you give some buzz you will allow me to train other LORAs or MODELS. Downloading only is not useful. For model visibility, remember to rate the images in my gallery and leave a like. Thank you

Version 2a is a further development of Version 1, but with a bit more freedom in the Design. It moves away somewhat from the Anime Style.

Version 2b is almost identical to Version 2. The difference between the two is that in

Version 2b, Eira Kishida once again uses the Design from the Anime.

Version 3 is an evolution of Version 2B and contains new prompts.

Version 3B is an alternative to Version 3. It is experimental.

Version 4, the trainings Data is bigger (more Pictures and Prompts used)

Version 5, the trainings Data is from Version 4 but the Prompts/Pictures are improved.

Version 6, the trainings Data is improved

Version 1 (Anima), trained with Anima and Prompts are improved

Recommended Settings Version 5:

Lora: 0.9-1.0
CFG Scale 3.5-6.0
Steps: 24-50

Sampler: Euler a

Trained with AdamW8Bit

Engine: Kohya

Recommended Settings Version 6:

Lora: 0.9-1.0
CFG Scale 3.5-5.0
Steps: 24-50

Sampler: Euler a, Trained with AdamW8Bit
Engine Kohya