Stable Diffusion V1 5

🚀 Stable Diffusion v1-5

Stable Diffusion is a latent text-to-image diffusion model. It can generate photo-realistic images from any text input, offering great potential for various creative and research applications.

🚀 Quick Start

This repository is being re-uploaded to HuggingFace under The CreativeML OpenRAIL-M License, specifically Section II which grants:

...a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable copyright license to reproduce, prepare, publicly display, publicly perform, sublicense, and distribute the Complementary Material, the Model, and Derivatives of the Model.

Note that these files originated from modelscope, not HuggingFace. Some files from the original repository may be missing. File integrity has been verified via checksum.

✨ Features

• Text-to-Image Generation: Capable of generating photo-realistic images based on text input.
• Latent Diffusion Model: Utilizes a latent diffusion model combined with an autoencoder for efficient image generation.
• Multiple Checkpoints: Six different checkpoints are provided, trained under various conditions.

📦 Installation

You can use this model with the 🧨Diffusers library. Install the necessary libraries via the following command:

pip install diffusers torch

💻 Usage Examples

Basic Usage

from diffusers import StableDiffusionPipeline
import torch

pipe = StableDiffusionPipeline.from_pretrained(
    "runwayml/stable-diffusion-v1-5",
    torch_dtype=torch.float16
)
pipe = pipe.to("cuda")

prompt = "a photo of an astronaut riding a horse on mars"
image = pipe(prompt).images[0]  
    
image.save("astronaut_rides_horse.png")

Advanced Usage

For more detailed instructions, use-cases, and examples in JAX, follow the instructions here.

📚 Documentation

Model Details

Uses

Direct Use

The model is for research purposes only. Possible research areas and tasks include:

• Safe deployment of models with the potential to generate harmful content.
• Probing and understanding the limitations and biases of generative models.
• Generation of artworks and use in design and other artistic processes.
• Applications in educational or creative tools.
• Research on generative models.

Misuse, Malicious Use, and Out-of-Scope Use

The model should not be used to intentionally create or disseminate images that create hostile or alienating environments for people. This includes generating disturbing, distressing, or offensive images, or content that propagates stereotypes.

Out-of-Scope Use

The model was not trained to provide factual or true representations of people or events. Using it for such content is beyond its capabilities.

Misuse and Malicious Use

Using the model to generate cruel content towards individuals is a misuse. This includes, but is not limited to:

• Generating demeaning, dehumanizing, or harmful representations of people, their environments, cultures, or religions.
• Intentionally promoting or propagating discriminatory content or harmful stereotypes.
• Impersonating individuals without consent.
• Generating sexual content without the consent of those who might see it.
• Spreading mis- and disinformation.
• Representing egregious violence and gore.
• Sharing copyrighted or licensed material in violation of its terms of use.
• Sharing altered copyrighted or licensed material in violation of its terms of use.

Limitations and Bias

Limitations

• The model does not achieve perfect photorealism.
• It cannot render legible text.
• It performs poorly on tasks involving compositionality, such as rendering “A red cube on top of a blue sphere”.
• Faces and people may not be generated properly.
• It was mainly trained with English captions and works less well in other languages.
• The autoencoding part of the model is lossy.
• It was trained on LAION-5B, which contains adult material and requires additional safety mechanisms for product use.
• There was no deduplication of the dataset, resulting in some memorization of duplicated training images. The training data can be searched at https://rom1504.github.io/clip-retrieval/.

Bias

Image generation models can reinforce or exacerbate social biases. Stable Diffusion v1 was trained on subsets of LAION-2B(en), mainly with English descriptions. Texts and images from non-English communities are likely underrepresented, leading to a default of white and western cultures in the output. The model's performance with non-English prompts is significantly worse.

Safety Module

The model is intended to be used with the Safety Checker in Diffusers. This checker compares model outputs against known hard-coded NSFW concepts in the embedding space of the CLIPTextModel after image generation.

Training

Training Data

The model was trained on the following dataset:

• LAION-2B (en) and subsets thereof.

Training Procedure

Stable Diffusion v1-5 is a latent diffusion model combining an autoencoder with a diffusion model trained in the autoencoder's latent space. During training:

• Images are encoded into latent representations by an encoder with a relative downsampling factor of 8, mapping H x W x 3 images to H/f x W/f x 4 latents.
• Text prompts are encoded by a ViT-L/14 text-encoder.
• The non-pooled output of the text encoder is fed into the UNet backbone of the latent diffusion model via cross-attention.
• The loss is a reconstruction objective between the added noise and the UNet's prediction.

Six Stable Diffusion checkpoints are provided, trained as follows:

• stable-diffusion-v1-1: 237,000 steps at 256x256 on laion2B-en, then 194,000 steps at 512x512 on laion-high-resolution.

• stable-diffusion-v1-2: Resumed from stable-diffusion-v1-1, 515,000 steps at 512x512 on "laion-improved-aesthetics".

• stable-diffusion-v1-3: Resumed from stable-diffusion-v1-2, 195,000 steps at 512x512 on "laion-improved-aesthetics" with 10% text-conditioning dropping.

• stable-diffusion-v1-4: Resumed from stable-diffusion-v1-2, 225,000 steps at 512x512 on "laion-aesthetics v2 5+" with 10% text-conditioning dropping.

• stable-diffusion-v1-5: Resumed from stable-diffusion-v1-2, 595,000 steps at 512x512 on "laion-aesthetics v2 5+" with 10% text-conditioning dropping.

• stable-diffusion-v1-5-inpainting: Resumed from stable-diffusion-v1-5, 440,000 steps of inpainting training at 512x512 on “laion-aesthetics v2 5+” with 10% text-conditioning dropping.

• Hardware: 32 x 8 x A100 GPUs

• Optimizer: AdamW

• Gradient Accumulations: 2

• Batch: 32 x 8 x 2 x 4 = 2048

• Learning rate: Warmup to 0.0001 for 10,000 steps and then kept constant

Evaluation Results

Evaluations with different classifier-free guidance scales (1.5, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0) and 50 PNDM/PLMS sampling steps show the relative improvements of the checkpoints:

Evaluated using 50 PLMS steps and 10000 random prompts from the COCO2017 validation set at 512x512 resolution. Not optimized for FID scores.

Environmental Impact

Stable Diffusion v1 Estimated Emissions

Based on the information, we estimate the following CO2 emissions using the Machine Learning Impact calculator from Lacoste et al. (2019).

• Hardware Type: A100 PCIe 40GB
• Hours used: 150000
• Cloud Provider: AWS
• Compute Region: US-east
• Carbon Emitted: 11250 kg CO2 eq.

Citation

    @InProceedings{Rombach_2022_CVPR,
        author    = {Rombach, Robin and Blattmann, Andreas and Lorenz, Dominik and Esser, Patrick and Ommer, Bj\"orn},
        title     = {High-Resolution Image Synthesis With Latent Diffusion Models},
        booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
        month     = {June},
        year      = {2022},
        pages     = {10684-10695}
    }

This model card was written by: Robin Rombach and Patrick Esser and is based on the DALL-E Mini model card.