Update README.md
Browse files
README.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
---
|
| 2 |
license: creativeml-openrail-m
|
| 3 |
-
thumbnail: "https://huggingface.co/
|
| 4 |
language:
|
| 5 |
- en
|
| 6 |
pipeline_tag: text-to-image
|
|
@@ -19,13 +19,13 @@ widget:
|
|
| 19 |
falling leaves, garden
|
| 20 |
example_title: example 1boy
|
| 21 |
datasets:
|
| 22 |
-
-
|
| 23 |
library_name: diffusers
|
| 24 |
---
|
| 25 |
|
| 26 |
# Anything V3.1
|
| 27 |
|
| 28 |
-
![Anime Girl](https://huggingface.co/
|
| 29 |
|
| 30 |
Anything V3.1 is a third-party continuation of a latent diffusion model, Anything V3.0. This model is claimed to be a better version of Anything V3.0 with a fixed VAE model and a fixed CLIP position id key. The CLIP reference was taken from Stable Diffusion V1.5. The VAE was swapped using Kohya's merge-vae script and the CLIP was fixed using Arena's stable-diffusion-model-toolkit webui extensions.
|
| 31 |
|
|
@@ -46,7 +46,7 @@ e.g. **_1girl, white hair, golden eyes, beautiful eyes, detail, flower meadow, c
|
|
| 46 |
- **Finetuned from model:** Anything V3.1
|
| 47 |
|
| 48 |
## How-to-Use
|
| 49 |
-
- Download `Anything V3.1` [here](https://huggingface.co/
|
| 50 |
- You need to adjust your prompt using aesthetic tags to get better result, you can use any generic negative prompt or use the following suggested negative prompt to guide the model towards high aesthetic generationse:
|
| 51 |
```
|
| 52 |
lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry
|
|
@@ -98,9 +98,9 @@ This model is overfitted and cannot follow prompts well, even after the text enc
|
|
| 98 |
|
| 99 |
Here is some cherrypicked samples and comparison between available models
|
| 100 |
|
| 101 |
-
![Anime Girl](https://huggingface.co/
|
| 102 |
-
![Anime Boy](https://huggingface.co/
|
| 103 |
-
![Aesthetic](https://huggingface.co/
|
| 104 |
|
| 105 |
## License
|
| 106 |
|
|
|
|
| 1 |
---
|
| 2 |
license: creativeml-openrail-m
|
| 3 |
+
thumbnail: "https://huggingface.co/cag/anything-v3-1/resolve/main/example-images/thumbnail.png"
|
| 4 |
language:
|
| 5 |
- en
|
| 6 |
pipeline_tag: text-to-image
|
|
|
|
| 19 |
falling leaves, garden
|
| 20 |
example_title: example 1boy
|
| 21 |
datasets:
|
| 22 |
+
- cag/anything-v3-1-dataset
|
| 23 |
library_name: diffusers
|
| 24 |
---
|
| 25 |
|
| 26 |
# Anything V3.1
|
| 27 |
|
| 28 |
+
|
| 29 |
|
| 30 |
Anything V3.1 is a third-party continuation of a latent diffusion model, Anything V3.0. This model is claimed to be a better version of Anything V3.0 with a fixed VAE model and a fixed CLIP position id key. The CLIP reference was taken from Stable Diffusion V1.5. The VAE was swapped using Kohya's merge-vae script and the CLIP was fixed using Arena's stable-diffusion-model-toolkit webui extensions.
|
| 31 |
|
|
|
|
| 46 |
- **Finetuned from model:** Anything V3.1
|
| 47 |
|
| 48 |
## How-to-Use
|
| 49 |
+
- Download `Anything V3.1` [here](https://huggingface.co/cag/anything-v3-1/resolve/main/anything-v3-1.safetensors), or `Anything V3.2` [here](https://huggingface.co/cag/anything-v3-1/resolve/main/anything-v3-2.safetensors), all model are in `.safetensors` format.
|
| 50 |
- You need to adjust your prompt using aesthetic tags to get better result, you can use any generic negative prompt or use the following suggested negative prompt to guide the model towards high aesthetic generationse:
|
| 51 |
```
|
| 52 |
lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry
|
|
|
|
| 98 |
|
| 99 |
Here is some cherrypicked samples and comparison between available models
|
| 100 |
|
| 101 |
+
|
| 102 |
+
|
| 103 |
+
|
| 104 |
|
| 105 |
## License
|
| 106 |
|