SegMoE-SD-4x2-v0: Segmind Mixture of Diffusion Experts

SegMoE-SD-4x2-v0 is an untrained Segmind Mixture of Diffusion Experts Model generated using segmoe from 4 Expert SD1.5 models. SegMoE is a powerful framework for dynamically combining Stable Diffusion Models into a Mixture of Experts within minutes without training. The framework allows for creation of larger models on the fly which offer larger knowledge, better adherence and better image quality.

Usage

This model can be used via the segmoe library.

Make sure to install segmoe by running

pip install segmoe

from segmoe import SegMoEPipeline

pipeline = SegMoEPipeline("segmind/SegMoE-SD-4x2-v0", device = "cuda")

prompt = "cosmic canvas, orange city background, painting of a chubby cat"
negative_prompt = "nsfw, bad quality, worse quality"
img = pipeline(
    prompt=prompt,
    negative_prompt=negative_prompt,
    height=1024,
    width=1024,
    num_inference_steps=25,
    guidance_scale=7.5,
).images[0]
img.save("image.png")

Config

Config Used to create this Model is:

base_model: SG161222/Realistic_Vision_V6.0_B1_noVAE
num_experts: 4
moe_layers: all
num_experts_per_tok: 2
experts:
  - source_model: SG161222/Realistic_Vision_V6.0_B1_noVAE
    positive_prompt: "cinematic, portrait, photograph, instagram, fashion, movie, macro shot, 8K, RAW, hyperrealistic, ultra realistic,"
    negative_prompt: " (deformed iris, deformed pupils, semi-realistic, cgi, 3d, render, sketch, cartoon, drawing, anime), text, cropped, out of frame, worst quality, low quality, jpeg artifacts, ugly, duplicate, morbid, mutilated, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, mutation, deformed, blurry, dehydrated, bad anatomy, bad proportions, extra limbs, cloned face, disfigured, gross proportions, malformed limbs, missing arms, missing legs, extra arms, extra legs, fused fingers, too many fingers, long neck"
  - source_model: dreamlike-art/dreamlike-anime-1.0
    positive_prompt: "photo anime, masterpiece, high quality, absurdres, 1girl, 1boy, waifu, chibi"
    negative_prompt: "simple background, duplicate, retro style, low quality, lowest quality, 1980s, 1990s, 2000s, 2005 2006 2007 2008 2009 2010 2011 2012 2013, bad anatomy, bad proportions, extra digits, lowres, username, artist name, error, duplicate, watermark, signature, text, extra digit, fewer digits, worst quality, jpeg artifacts, blurry"
  - source_model: Lykon/dreamshaper-8
    positive_prompt: "bokeh, intricate, elegant, sharp focus, soft lighting, vibrant colors, dreamlike, fantasy, artstation, concept art"
    negative_prompt: "low quality, lowres, jpeg artifacts, signature, bad anatomy, extra legs, extra arms, extra fingers, poorly drawn hands, poorly drawn feet, disfigured, out of frame, tiling, bad art, deformed, mutated, blurry, fuzzy, misshaped, mutant, gross, disgusting, ugly, watermark, watermarks"
  - source_model: dreamlike-art/dreamlike-diffusion-1.0
    positive_prompt: "dreamlikeart, a grungy woman with rainbow hair, travelling between dimensions, dynamic pose, happy, soft eyes and narrow chin, extreme bokeh, dainty figure, long hair straight down, torn kawaii shirt and baggy jeans, In style of by Jordan Grimmer and greg rutkowski, crisp lines and color, complex background, particles, lines, wind, concept art, sharp focus, vivid colors"
    negative_prompt: "nude, naked, low quality, lowres, jpeg artifacts, signature, bad anatomy, extra legs, extra arms, extra fingers, poorly drawn hands, poorly drawn feet, disfigured, out of frame"

Other Variants

We release 3 merges on Hugging Face,

• SegMoE 4x2 has four Stable Diffusion XL expert models.
• SegMoE 2x1 has two Stable Diffusion XL expert models.

Model Description

• Developed by: Segmind
• Developers: Yatharth Gupta and Vishnu Jaddipal.
• Model type: Diffusion-based text-to-image generative mixture of experts model
• License: Apache 2.0

Out-of-Scope Use

The SegMoE-SD-4x2-v0 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.

Advantages

• Benefits from The Knowledge of Several Finetuned Experts
• Training Free
• Better Adaptability to Data
• Model Can be upgraded by using a better finetuned model as one of the experts.

Limitations

• Though the Model improves upon the fidelity of images as well as adherence, it does not be drastically better than any one expert without training and relies on the knowledge of the experts.
• This is not yet optimized for speed.
• The framework is not yet optimized for memory usage.

Citation

@misc{segmoe,
  author = {Yatharth Gupta, Vishnu V Jaddipal, Harish Prabhala},
  title = {SegMoE},
  year = {2024},
  publisher = {HuggingFace},
  journal = {HuggingFace Models},
  howpublished = {\url{https://huggingface.co./segmind/SegMoE-SD-4x2-v0}}
}