README.md

---
license: creativeml-openrail-m
language:
- en
tags:
- art
- Stable Diffusion
---
## Model Card for lyraSD2

lyraSD2 is currently the **fastest Stable Diffusion model** that can 100% align the outputs of **diffusers** available, boasting an inference cost of only **0.52 seconds** for a 512x512 image, accelerating the process up to **80% faster** than the original version. 

Among its main features are:

- 4 Commonly used Pipelines
- - Text2Img
- - Img2Img
- - ControlNetText2Img
- - ControlNetImg2Img
- 100% likeness to diffusers output
- ControlNet Hot Swap: Can hot swap a ControlNet model weights within 0.4s (0s if cached)
- Lora How Swap: Can hot swap a Lora within 0.5s (0.1s if cached)
- device requirements: Nvidia Ampere architecture (A100, A10) or compatable

## Speed

### test environment

- device: Nvidia A100 40G
- img size: 512x512
- percision:fp16
- steps: 20
- sampler: EulerA

### Text2Img
|model|time cost(ms)|
|:-:|:-:|
|torch2.0.1 + diffusers|~667ms|
|lyraSD|~528ms|

### ControlNet-Text2Img
|model|time cost(ms)|
|:-:|:-:|
|torch2.0.1 + diffusers|~930ms|
|lyraSD2|~745ms|

## Model Sources

- **Checkpoint:** https://civitai.com/models/7371/rev-animated
- **ControlNet:** https://huggingface.co/lllyasviel/sd-controlnet-canny
- **Lora:** https://civitai.com/models/18323?modelVersionId=46846

## Text2Img Uses

```python
import torch
import time

from lyrasd_model import LyraSdTxt2ImgPipeline

# 存放模型文件的路径，应该包含一下结构：
#   1. clip 模型
#   2. 转换好的优化后的 unet 模型，放入其中的 unet_bins 文件夹
#   3. vae 模型
#   4. scheduler 配置

# LyraSD 的 C++ 编译动态链接库，其中包含 C++ CUDA 计算的细节
lib_path = "./lyrasd_model/lyrasd_lib/libth_lyrasd_cu11_sm80.so"
model_path = "./models/lyrasd_rev_animated"
lora_path = "./models/lyrasd_xiaorenshu_lora"

# 构建 Txt2Img 的 Pipeline
model = LyraSdTxt2ImgPipeline(model_path, lib_path)

# load lora
# 参数分别为 lora 存放位置，名字，lora 强度，lora模型精度
model.load_lora(lora_path, "xiaorenshu", 0.4, "fp32")

# 准备应用的输入和超参数
prompt = "a cat, cute, cartoon, concise, traditional, chinese painting, Tang and Song Dynasties, masterpiece, 4k, 8k, UHD, best quality"
negative_prompt = "(((horrible))), (((scary))), (((naked))), (((large breasts))), high saturation, colorful, human:2, body:2, low quality, bad quality, lowres, out of frame, duplicate, watermark, signature, text, frames, cut, cropped, malformed limbs, extra limbs, (((missing arms))), (((missing legs)))"
height, width = 512, 512
steps = 30
guidance_scale = 7
generator = torch.Generator().manual_seed(123)
num_images = 1

start = time.perf_counter()
# 推理生成
images = model(prompt, height, width, steps,
        guidance_scale, negative_prompt, num_images,
        generator=generator)
print("image gen cost: ",time.perf_counter() - start)
# 存储生成的图片
for i, image in enumerate(images):
    image.save(f"outputs/res_txt2img_lora_{i}.png")

# unload lora，参数为 lora 的名字，是否清除 lora 缓存
# model.unload_lora("xiaorenshu", True)


```
## Demo output

### Text2Img
#### Text2Img without Lora
![text2img_demo](./outputs/res_txt2img_0.png)

#### Text2Img with Lora
![text2img_demo](./outputs/res_txt2img_lora_0.png)

### Img2Img

#### Img2Img input
<img src="https://chuangxin-research-1258344705.cos.ap-guangzhou.myqcloud.com/share/files/seaside_town.png?q-sign-algorithm=sha1&q-ak=AKIDBF6i7GCtKWS8ZkgOtACzX3MQDl37xYty&q-sign-time=1692601590;1865401590&q-key-time=1692601590;1865401590&q-header-list=&q-url-param-list=&q-signature=ca04ca92d990d94813029c0d9ef29537e5f4637c" alt="img2img input" width="512"/>

#### Img2Img output
![text2img_demo](./outputs/res_img2img_0.png)

### ControlNet Text2Img

#### Control Image
![text2img_demo](./control_bird_canny.png)

#### ControlNet Text2Img Output
![text2img_demo](./outputs/res_controlnet_txt2img_0.png)

## Docker Environment Recommendation

- For Cuda 11.X: we recommend ```nvcr.io/nvidia/pytorch:22.12-py3```
- For Cuda 12.0: we recommend ```nvcr.io/nvidia/pytorch:23.02-py3```

```bash
docker pull nvcr.io/nvidia/pytorch:23.02-py3
docker run --rm -it --gpus all -v ./:/lyraSD2 nvcr.io/nvidia/pytorch:23.02-py3

pip install -r requirements.txt
python txt2img_demo.py
```

## Citation
``` bibtex
@Misc{lyraSD2_2023,
  author =       {Kangjian Wu, Zhengtao Wang, Yibo Lu, Haoxiong Su, Bin Wu},
  title =        {lyraSD2: Accelerating Stable Diffusion with best flexibility},
  howpublished = {\url{https://huggingface.co/TMElyralab/lyraSD2}},
  year =         {2023}
}
```

## Report bug
- start a discussion to report any bugs!--> https://huggingface.co/TMElyralab/lyraSD2/discussions
- report bug with a `[bug]` mark in the title.