File size: 2,289 Bytes
388c240 76edebb 1bef545 bea3a14 76edebb db32dc3 76edebb | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 | ---
license: other
license_name: circlestone-labs-non-commercial-license
license_link: https://huggingface.co/circlestone-labs/Anima
base_model:
- circlestone-labs/Anima
pipeline_tag: text-to-image
---
<style>
.gallery {
display: grid;
grid-template-columns: repeat(3, 1fr);
gap: 10px;
}
.gallery img {
width: 100%;
height: auto;
margin-top: 0px;
margin-bottom: 0px;
border-radius: 10px;
box-shadow: 0 4px 8px rgba(0, 0, 0, 0.2);
transition: transform 0.3s;
}
</style>
# Showcase
<div class="gallery">
<img src="./sample01.png" alt="sample01">
<img src="./sample02.png" alt="sample02">
<img src="./sample03.png" alt="sample03">
</div>
<br>
## Notice: v1.1 Update
`Nameless-Anima11.safetensors` has been added as version 1.1.
This version was created by rebuilding the Style LoRA used for the model merge.
The previous version, `Nameless-Anima10.safetensors`, will remain available.
### Main Changes
* Rebuilt the Style LoRA from scratch.
* Merged the new Style LoRA to create v1.1.
* Improved overall style stability.
* Output may differ slightly from v1.0 in line quality, color tone, texture, and character rendering.
If you prefer the previous output style, please use `Nameless-Anima10.safetensors`.
### Style LoRA Training Details
The Style LoRA used for v1.1 was trained on a dataset of 500 images generated with my custom merged model, [NONAMEmix-Vpred_v3](https://huggingface.co/Emanon14/NONAMEmix-Vpred_v3).
Training settings:
* Learning rate: `4e-5`
* Self-Attention learning rate: `0`
* Cross-Attention learning rate: `0`
* Rank: `4`
* Alpha: `4`
* Batch size: `2`
* Gradient accumulation steps: `2`
* Training steps: `1,000`
<br>
# Model Description
This model was created by merging a custom LoRA into the base model, **anima-base-v1.0**.
The LoRA was trained on a small dataset in order to stabilize the generation style.
## Usage & Generation Settings
For details regarding generation settings, prompting guidelines, and the use of various tags, please refer to the original [Anima model documentation](https://huggingface.co/circlestone-labs/Anima).
## License
The license for this model strictly follows the license of the base model, Anima. Please refer to the original Anima license for exact terms and conditions of use. |