|
|
--- |
|
|
license: mit |
|
|
language: |
|
|
- en |
|
|
new_version: SkillForge/GPUDiff-V1 |
|
|
base_model: |
|
|
- SkillForge/GPUDiff-V1 |
|
|
pipeline_tag: text-to-image |
|
|
library_name: transformers |
|
|
tags: |
|
|
- art |
|
|
--- |
|
|
## GPUDiff-V1 - good model for image generation! |
|
|
GPUDiff - this is a open-source model, but the model is licensed, be careful. |
|
|
## Usage |
|
|
```bash |
|
|
python -c "from model import train; train()" # train the model |
|
|
``` |
|
|
```bash |
|
|
python -c "from model import demo.launch(); demo.launch()" # launch the demo diffusion model |
|
|
``` |
|
|
## Features |
|
|
**Fast generation** - no longer generations on a powerful models. |
|
|
|
|
|
**No Mega Powerful Model** - brilliant settings of GPU. |
|
|
|
|
|
**Weighs little** - just a one file. |
|
|
## Coco Train |
|
|
This model is training on a Coco Dataset 2017. |
|
|
Using Coco Urls: |
|
|
|
|
|
Coco images: http://images.cocodataset.org/zips/train2017.zip |
|
|
|
|
|
Coco annotations: http://images.cocodataset.org/annotations/annotations_trainval2017.zip |
|
|
## License |
|
|
This is model is a lecensed, see the LICENSE for more information. |
|
|
## Creator |
|
|
Creator is a me (EidolonAI, SkillForge45). |
