metadata

license: apache-2.0
tags:
  - text-to-image
  - ultra-realistic
  - text-to-image
  - stable-diffusion
  - distilled-model
  - knowledge-distillation
pinned: true
datasets:
  - zzliang/GRIT
  - wanng/midjourney-v5-202304-clean
library_name: diffusers

Segmind Stable Diffusion 1B (SSD-1B) Model Card

Model Description

The Segmind Stable Diffusion Model (SSD-1B) is a distilled 50% smaller version of the Stable Diffusion XL (SDXL), offering a 60% speedup while maintaining high-quality text-to-image generation capabilities. It has been trained on diverse datasets, including Grit and Midjourney scrape data, to enhance its ability to create a wide range of visual content based on textual prompts.

This model employs a knowledge distillation strategy, where it leverages the teachings of several expert models in succession, including SDXL, ZavyChromaXL, and JuggernautXL, to combine their strengths and produce impressive visual outputs.

Special thanks to the HF team 🤗 especially Sayak, Patrick and Poli for their collaboration and guidance on this work.

Demo

Try out the model at Segmind SSD-1B for ⚡ fastest inference. You can also try it on 🤗 Spaces

Image Comparision (SDXL-1.0 vs SSD-1B)

Usage:

This model can be used via the 🧨 Diffusers library.

Make sure to install diffusers from source by running

pip install git+https://github.com/huggingface/diffusers

In addition, please install transformers, safetensors and accelerate:

pip install transformers accelerate safetensors

To use the model, you can run the following:

from diffusers import StableDiffusionXLPipeline
import torch
pipe = StableDiffusionXLPipeline.from_pretrained("segmind/SSD-1B", torch_dtype=torch.float16, use_safetensors=True, variant="fp16")
pipe.to("cuda")
# if using torch < 2.0
# pipe.enable_xformers_memory_efficient_attention()
prompt = "An astronaut riding a green horse" # Your prompt here
neg_prompt = "ugly, blurry, poor quality" # Negative prompt here
image = pipe(prompt=prompt, negative_prompt=neg_prompt).images[0]

Model Description

Developed by: Segmind
Developers: Yatharth Gupta and Vishnu Jaddipal.
Model type: Diffusion-based text-to-image generative model
License: Apache 2.0
Distilled From stabilityai/stable-diffusion-xl-base-1.0

Please do use negative prompting for the best quality!

Key Features

Text-to-Image Generation: The model excels at generating images from text prompts, enabling a wide range of creative applications.
Distilled for Speed: Designed for efficiency, this model offers a 60% speedup, making it a practical choice for real-time applications and scenarios where rapid image generation is essential.
Diverse Training Data: Trained on diverse datasets, the model can handle a variety of textual prompts and generate corresponding images effectively.
Knowledge Distillation: By distilling knowledge from multiple expert models, the Segmind Stable Diffusion Model combines their strengths and minimizes their limitations, resulting in improved performance.

Model Architecture

The SSD-1B Model is a 1.3B Parameter Model which has several layers removed from the Base SDXL Model

Training info

These are the key hyperparameters used during training:

Steps: 251000
Learning rate: 1e-5
Batch size: 32
Gradient accumulation steps: 4
Image resolution: 1024
Mixed-precision: fp16

Speed Comparision

We have observed that SSD-1B is upto 60% faster than the Base SDXL Model. Below is a comparision on an A100 40GB.

Model Sources

For research and development purposes, the SSD-1B Model can be accessed via the Segmind AI platform. For more information and access details, please visit Segmind.

Uses

Direct Use

The Segmind Stable Diffusion Model is suitable for research and practical applications in various domains, including:

Art and Design: It can be used to generate artworks, designs, and other creative content, providing inspiration and enhancing the creative process.
Education: The model can be applied in educational tools to create visual content for teaching and learning purposes.
Research: Researchers can use the model to explore generative models, evaluate its performance, and push the boundaries of text-to-image generation.
Safe Content Generation: It offers a safe and controlled way to generate content, reducing the risk of harmful or inappropriate outputs.
Bias and Limitation Analysis: Researchers and developers can use the model to probe its limitations and biases, contributing to a better understanding of generative models' behavior.

Downstream Use

The Segmind Stable Diffusion Model can also be used directly with the 🧨 Diffusers library training scripts for further training, including:

Fine-Tune:

export MODEL_NAME="segmind/SSD-1B"
export VAE_NAME="madebyollin/sdxl-vae-fp16-fix"
export DATASET_NAME="lambdalabs/pokemon-blip-captions"

accelerate launch train_text_to_image_lora_sdxl.py \
  --pretrained_model_name_or_path=$MODEL_NAME \
  --pretrained_vae_model_name_or_path=$VAE_NAME \
  --dataset_name=$DATASET_NAME --caption_column="text" \
  --resolution=1024 --random_flip \
  --train_batch_size=1 \
  --num_train_epochs=2 --checkpointing_steps=500 \
  --learning_rate=1e-04 --lr_scheduler="constant" --lr_warmup_steps=0 \
  --mixed_precision="fp16" \
  --seed=42 \
  --output_dir="sd-pokemon-model-lora-sdxl" \
  --validation_prompt="cute dragon creature" --report_to="wandb" \
  --push_to_hub

LoRA:

export MODEL_NAME="segmind/SSD-1B"
export VAE_NAME="madebyollin/sdxl-vae-fp16-fix"
export DATASET_NAME="lambdalabs/pokemon-blip-captions"

accelerate launch train_text_to_image_sdxl.py \
  --pretrained_model_name_or_path=$MODEL_NAME \
  --pretrained_vae_model_name_or_path=$VAE_NAME \
  --dataset_name=$DATASET_NAME \
  --enable_xformers_memory_efficient_attention \
  --resolution=512 --center_crop --random_flip \
  --proportion_empty_prompts=0.2 \
  --train_batch_size=1 \
  --gradient_accumulation_steps=4 --gradient_checkpointing \
  --max_train_steps=10000 \
  --use_8bit_adam \
  --learning_rate=1e-06 --lr_scheduler="constant" --lr_warmup_steps=0 \
  --mixed_precision="fp16" \
  --report_to="wandb" \
  --validation_prompt="a cute Sundar Pichai creature" --validation_epochs 5 \
  --checkpointing_steps=5000 \
  --output_dir="sdxl-pokemon-model" \
  --push_to_hub

Dreambooth LoRA:

export MODEL_NAME="segmind/SSD-1B"
export INSTANCE_DIR="dog"
export OUTPUT_DIR="lora-trained-xl"
export VAE_PATH="madebyollin/sdxl-vae-fp16-fix"

accelerate launch train_dreambooth_lora_sdxl.py \
  --pretrained_model_name_or_path=$MODEL_NAME  \
  --instance_data_dir=$INSTANCE_DIR \
  --pretrained_vae_model_name_or_path=$VAE_PATH \
  --output_dir=$OUTPUT_DIR \
  --mixed_precision="fp16" \
  --instance_prompt="a photo of sks dog" \
  --resolution=1024 \
  --train_batch_size=1 \
  --gradient_accumulation_steps=4 \
  --learning_rate=1e-5 \
  --report_to="wandb" \
  --lr_scheduler="constant" \
  --lr_warmup_steps=0 \
  --max_train_steps=500 \
  --validation_prompt="A photo of sks dog in a bucket" \
  --validation_epochs=25 \
  --seed="0" \
  --push_to_hub

Out-of-Scope Use

The SSD-1B Model is not suitable for creating factual or accurate representations of people, events, or real-world information. It is not intended for tasks requiring high precision and accuracy.

Limitations and Bias

Limitations

Photorealism: The model does not achieve perfect photorealism and may produce images with artistic or stylized qualities.
Legible Text: Generating legible text within images is a challenge for the model, and text within images may appear distorted or unreadable.
Compositionality: Complex tasks involving composition, such as rendering images based on intricate descriptions, may pose challenges for the model.
Faces and People: While the model can generate a wide range of content, it may not consistently produce realistic or high-quality images of faces and people.
Lossy Autoencoding: The autoencoding aspect of the model is lossy, which means that some details in the input text may not be perfectly retained in the generated images.

Bias

The SSD-1B Model is trained on a diverse dataset, but like all generative models, it may exhibit biases present in the training data. Users are encouraged to be mindful of potential biases in the model's outputs and take appropriate steps to mitigate them.