X Stable Diffusion Save

Welcome to x-stable-diffusion by Stochastic!

This project is a compilation of acceleration techniques for the Stable Diffusion model to help you generate images faster and more efficiently, saving you both time and money.

With example images and a comprehensive benchmark, you can easily choose the best technique for your needs. When you're ready to deploy, our CLI called stochasticx makes it easy to get started on your local machine. Try x-stable-diffusion and see the difference it can make for your image generation performance and cost savings.

🚀 Installation

Quickstart

Make sure you have Python and Docker installed on your system

1. Install the latest version of stochasticx library.

pip install stochasticx

2. Deploy the Stable Diffusion model

stochasticx stable-diffusion deploy --type aitemplate

Alternatively, you can deploy stable diffusion without our CLI by checking the steps here.

3. To perform inference with this deployed model:

stochasticx stable-diffusion inference --prompt "Riding a horse"

Check all the options of the inference command:

stochasticx stable-diffusion inference --help

4. You can get the logs of the deployment executing the following command:

stochasticx stable-diffusion logs

5. Stop and remove the deployment with this command:

stochasticx stable-diffusion stop

How to get less than 1s latency?

Change the num_inference_steps to 30. With this, you can get an image generated in 0.88 seconds.

{
  'max_seq_length': 64,
  'num_inference_steps': 30, 
  'image_size': (512, 512) 
}

You can also experiment with reducing the image_size.

How to run on Google Colab?

• Try PyTorch - FP16 in Colab ->
• Try TensorRT in Colab ->

In each folder, we will provide a Google Colab notebook with which you can test the full flow and inference on a T4 GPU

Manual deployment

Check the README.md of the following directories:

• AITemplate
• FlashAttention
• nvFuser
• PyTorch
• TensorRT

🔥 Optimizations

• AITemplate: Latest optimization framework of Meta
• TensorRT: NVIDIA TensorRT framework
• nvFuser: nvFuser with Pytorch
• FlashAttention: FlashAttention intergration in Xformers

Benchmarks

Setup

For hardware, we used 1x40GB A100 GPU with CUDA 11.6 and the results are reported by averaging 50 runs.

The following arguments were used for image generation for all the benchmarks:

{
  'max_seq_length': 64,
  'num_inference_steps': 50, 
  'image_size': (512, 512) 
}

Online results

For batch_size 1, these are the latency results:

A100 GPU

project	Latency (s)	GPU VRAM (GB)
PyTorch fp16	5.77	10.3
nvFuser fp16	3.15	---
FlashAttention fp16	2.80	7.5
TensorRT fp16	1.68	8.1
AITemplate fp16	1.38	4.83
ONNX (CUDA)	7.26	13.3

T4 GPU

Note: AITemplate might not support T4 GPU yet. Check support here

project	Latency (s)
PyTorch fp16	16.2
nvFuser fp16	19.3
FlashAttention fp16	13.7
TensorRT fp16	9.3

Batched results - A100 GPU

The following results were obtained by varying batch_size from 1 to 24.

project \ bs	1	4	8	16	24
Pytorch fp16	5.77s/10.3GB	19.2s/18.5GB	36s/26.7GB	OOM
FlashAttention fp16	2.80s/7.5GB	9.1s/17GB	17.7s/29.5GB	OOM
TensorRT fp16	1.68s/8.1GB	OOM
AITemplate fp16	1.38s/4.83GB	4.25s/8.5GB	7.4s/14.5GB	15.7s/25GB	23.4s/36GB
ONNX (CUDA)	7.26s/13.3GB	OOM	OOM	OOM	OOM

Note: TensorRT fails to convert UNet model from ONNX to TensorRT due to memory issues.

Sample images generated

Click here to view the complete list of generated images

Optimization \ Prompt	Super Mario learning to fly in an airport, Painting by Leonardo Da Vinci	The Easter bunny riding a motorcycle in New York City	Drone flythrough of a tropical jungle convered in snow
PyTorch fp16
nvFuser fp16
FlashAttention fp16
TensorRT fp16
AITemplate fp16

References

• HuggingFace Diffusers
• AITemplate

🌎 Join our community

• Discord - https://discord.gg/TgHXuSJEk6

🌎 Contributing

As an open source project in a rapidly evolving field, we welcome contributions of all kinds, including new features and better documentation. Please read our contributing guide to learn how you can get involved.

For managed hosting on our cloud or on your private cloud [Contact us →](https://stochastic.ai/contact)