---
license: apache-2.0
pipeline_tag: text-to-image
tags:
- text-to-image
- stable-diffusion
- safetensors
- stable-diffusion-1.5
- moe
- segmoe
language:
- en
library_name: diffusers
---

## Warning
This is an experimental model. It works only with segmoe library!

## Experts
  - source_model: Lykon/dreamshaper-8 (base)
  - source_model: Lykon/AAM_AnyLora_AnimeMix
  - source_model: stablediffusionapi/realistic-vision-51

## Usage

This model can be used via the [segmoe](https://github.com/segmind/segmoe) library. 

Make sure to install segmoe by running

```bash
pip install segmoe
```

```python
from segmoe import SegMoEPipeline

pipeline = SegMoEPipeline("RachidAR/AFlow-SegMoe-1Bx3-v0.1", device = "cuda", safety_checker = None)

prompt = "cosmic canvas, orange city background, painting of a chubby cat"
negative_prompt = "nsfw, bad quality, worse quality"
img = pipeline(
    prompt=prompt,
    negative_prompt=negative_prompt,
    height=1024,
    width=1024,
    num_inference_steps=25,
    guidance_scale=7.5,
).images[0]
img.save("image.png")
```

![image/png](https://huggingface.co/RachidAR/AFlow-SegMoe-1Bx3-v0.1/resolve/main/example1.png)
![image/png](https://huggingface.co/RachidAR/AFlow-SegMoe-1Bx3-v0.1/resolve/main/example2.png)
![image/png](https://huggingface.co/RachidAR/AFlow-SegMoe-1Bx3-v0.1/resolve/main/example3.png)