---
license: creativeml-openrail-m
language:
- zh
- en
library_name: diffusers
pipeline_tag: text-to-image
tags:
- art
- tensorRT
---
## Model Card for lyraSD

lyraSD is a tensorRT implementation of Stable diffusion model, with stable 1.5 weights. Main features 

- weights: original stable diffusion 1.5 weights
- input image: 512x512 (in img2img mode)
- output image: 512x512
- device requirements: Nvidia Ampere architecture (A100) or compatable
- super-resultion: 4x by default, optional.

lyraSD is a part of TME lyralab Fast Model Plan.

## Speed

### test on A100, img size: 512x512，superresolution=False, mode=text2img
|mode|version|speed|
|:-:|:-:|:-:|
|text2img|TensorRT official|0.41s/img|
|text2img|LyraSD|0.36s/img|





## Model Sources [optional]

- **Repository:** [https://huggingface.co/runwayml/stable-diffusion-v1-5]

## Uses

```python
from lyraSD import LyraSD

t2imodel = LyraSD("text2img", "./sd1.5-engine")
t2imodel.inference(prompt="a red ballon flying in the sky")


from PIL import Image
i2imodel = LyraSD("img2img", "./sd1.5-engine")
demo_img = Image.open("output/text2img_demo.jpg")
i2imodel.inference(prompt="comic style", image=demo_img)
```

## environment

- hardware: Nvidia Ampere architecture (A100) or compatable
- docker image avaible: https://hub.docker.com/r/bigmoyan/lyra_aigc/tags
```
docker pull bigmoyan/lyra_aigc:v0.1
```

## report bug
- start a discussion to report any bugs!--> https://huggingface.co/TMElyralab/lyraSD/discussions
- report bug with a `[bug]` mark in the title.