Upload 2540444/2855073/README.md with huggingface_hub
Browse files- 2540444/2855073/README.md +50 -0
2540444/2855073/README.md
ADDED
|
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
tags:
|
| 4 |
+
- anime
|
| 5 |
+
- style
|
| 6 |
+
---
|
| 7 |
+
|
| 8 |
+
# Anima Highres/Aesthetic Boost - v1.0 - 2855073
|
| 9 |
+
|
| 10 |
+
**Model Type**: LORA
|
| 11 |
+
|
| 12 |
+
**Base Model**: Anima
|
| 13 |
+
|
| 14 |
+
**Trigger Words**: None
|
| 15 |
+
|
| 16 |
+
**Tags**: anime, style
|
| 17 |
+
|
| 18 |
+
## Gallery
|
| 19 |
+
|
| 20 |
+
<table>
|
| 21 |
+
<tr>
|
| 22 |
+
<td><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/91519397-ca49-4771-a3d6-c06bbae06b25/original=true/127347641.jpeg" width="200"></td>
|
| 23 |
+
<td><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/7e859de6-131c-40dd-9ddb-79c563d39340/original=true/127347643.jpeg" width="200"></td>
|
| 24 |
+
<td><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/ce282281-70a9-45b6-919a-3afc1c65ea9d/original=true/127347646.jpeg" width="200"></td>
|
| 25 |
+
</tr>
|
| 26 |
+
<tr>
|
| 27 |
+
<td><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/6f445b11-e116-48ea-ae4a-b0ad8b1f6b5e/original=true/127347642.jpeg" width="200"></td>
|
| 28 |
+
</tr>
|
| 29 |
+
</table>
|
| 30 |
+
|
| 31 |
+
## Description
|
| 32 |
+
|
| 33 |
+
Showcase images are base model (left) vs LoRA (right).
|
| 34 |
+
|
| 35 |
+
This LoRA does two things:
|
| 36 |
+
|
| 37 |
+
1. Allows generating at higher resolutions. 1536 works without any major issues, and even 2048 (4 MP) now works without completely falling apart.
|
| 38 |
+
2. Slight aesthetic increase toward higher-quality images.
|
| 39 |
+
|
| 40 |
+
It is trained on 10,000 high rated Danbooru images, with mixed-res training at [512, 1024, 1536] resolutions. The images are sampled to have a more even balance of SFW/NSFW and male/female (because otherwise, selecting on ratings only, you would get almost 100% NSFW female). The aesthetic boost is fairly weak, and is prompt dependent. It is not targeting any particular aesthetic, but simply "high rated Danbooru images", so there is still a lot of stylistic diversity, and hence the LoRA's effect is subtle.
|
| 41 |
+
|
| 42 |
+
---
|
| 43 |
+
|
| 44 |
+
Author: [circlestone_labs](https://civitai.com/user/circlestone_labs)
|
| 45 |
+
|
| 46 |
+
Model: [CivitAI Model Page](https://civitai.com/models/2540444?modelVersionId=2855073)
|
| 47 |
+
|
| 48 |
+
Archive: [CivArchive Page](https://civarchive.com/models/2540444?modelVersionId=2855073)
|
| 49 |
+
|
| 50 |
+
<!-- Version: 20260601_upload -->
|