Skip to content

Latest commit

 

History

History
 
 

stable_diffusion

README.md

Stable Diffusion

Stable Diffusion in MLX. The implementation was ported from Hugging Face's diffusers and we are fetching and using the weights available on the Hugging Face Hub by Stability AI at stabilitiai/stable-diffusion-2-1.

out
Image generated using Stable Diffusion in MLX and the prompt 'A big red sign saying MLX in capital letters.'

Installation

The dependencies are minimal, namely:

  • safetensors and huggingface-hub to load the checkpoints.
  • regex for the tokenization
  • numpy because safetensors needs to return some form of array
  • tqdm and PIL for the txt2image.py script

You can install all of the above with the requirements.txt as follows:

pip install -r requirements.txt

Usage

Although each component in this repository can be used by itself, the fastest way to get started is by using the StableDiffusion class from the stable_diffusion module.

import mlx.core as mx
from stable_diffusion import StableDiffusion

# This will download all the weights from HF hub and load the models in
# memory
sd = StableDiffusion()

# This creates a python generator that returns the latent produced by the
# reverse diffusion process.
#
# Because MLX is lazily evaluated iterating over this generator doesn't
# actually perform the computation until mx.eval() is called.
latent_generator = sd.generate_latents("A photo of an astronaut riding a horse on Mars.")

# Here we are evaluating each diffusion step but we could also evaluate
# once at the end.
for x_t in latent_generator:
    mx.eval(x_t)

# Now x_t is the last latent from the reverse process aka x_0. We can
# decode it into an image using the stable diffusion VAE.
im = sd.decode(x_t)

The above is almost line for line the implementation of the txt2image.py script in the root of the repository. You can use the script as follows:

python txt2image.py "A photo of an astronaut riding a horse on Mars." --n_images 4 --n_rows 2

Image 2 Image

There is also the option of generating images based on another image using the example script image2image.py. To do that an image is first encoded using the autoencoder to get its latent representation and then noise is added according to the forward diffusion process and the strength parameter. A strength of 0.0 means no noise and a strength of 1.0 means starting from completely random noise.

image2image
Generations with varying strength using the original image and the prompt 'A lit fireplace'.

The command to generate the above images is:

python image2image.py --strength 0.5 original.png 'A lit fireplace'

Note: image2image.py will automatically downsample your input image to guarantee that its dimensions are divisible by 64. If you want full control of this process, resize your image prior to using the script.

Performance

The following table compares the performance of the UNet in stable diffusion. We report throughput in images per second processed by the UNet for the provided txt2image.py script and the diffusers library using the MPS PyTorch backend.

At the time of writing this comparison convolutions are still some of the least optimized operations in MLX. Despite that, MLX still achieves ~40% higher throughput than PyTorch with a batch size of 16 and ~15% higher when comparing the optimal batch sizes.

Notably, PyTorch achieves almost ~50% higher throughput for the batch size of 1 which is unfortunate as that means that a single image can be computed faster. However, when starting with the models not loaded in memory and PyTorch's MPS graph kernels not cached, the compilation time more than accounts for this speed difference.

Batch size PyTorch MLX
1 6.25 im/s 4.17 im/s
2 7.14 im/s 5.88 im/s
4 7.69 im/s 7.14 im/s
6 7.22 im/s 8.00 im/s
8 6.89 im/s 8.42 im/s
12 6.62 im/s 8.51 im/s
16 6.32 im/s 8.79 im/s

The above experiments were made on an M2 Ultra with PyTorch version 2.1, diffusers version 0.21.4 and transformers version 4.33.3. For the generation we used classifier free guidance which means that the above batch sizes result double the images processed by the UNet.

Note that the above table means that it takes about 90 seconds to fully generate 16 images with MLX and 50 diffusion steps with classifier free guidance and about 120 for PyTorch.