terminusresearch/stable-diffusion-3.0-medium_reality-mix

sd3-reality-mix

This is a full rank finetune derived from stabilityai/stable-diffusion-3-medium-diffusers.

The main validation prompt used during training was:

ethnographic photography of teddy bear at a picnic holding a sign that says SOON, sitting next to a red sphere which is inside a capsule

Validation settings

• —CFG: 5.5
• —CFG Rescale: 0.0
• —Steps: 30
• —Sampler: euler
• —Seed: 42
• —Resolutions: 512x512,1024x1024,1280x768,960x1152

Note: The validation settings are not necessarily the same as the training settings.

You can find some example images in the following gallery:

<Gallery />

The text encoder was not trained. You may reuse the base model text encoder for inference.

Training settings

• —Training epochs: 0
• —Training steps: 7000
• —Learning rate: 1e-06
• —Effective batch size: 96
• —Micro-batch size: 6
• —Gradient accumulation steps: 2
• —Number of GPUs: 8
• —Prediction type: epsilon
• —Rescaled betas zero SNR: False
• —Optimizer: AdamW, stochastic bf16
• —Precision: Pure BF16
• —Xformers: Not used

Datasets

photo-concept-bucket

• —Repeats: 0
• —Total number of images: ~557568
• —Total number of aspect buckets: 38
• —Resolution: 1.0 megapixels
• —Cropped: False
• —Crop style: None
• —Crop aspect: None

dalle3

• —Repeats: 0
• —Total number of images: ~984960
• —Total number of aspect buckets: 44
• —Resolution: 1.0 megapixels
• —Cropped: False
• —Crop style: None
• —Crop aspect: None

Inference

import torch
from diffusers import StableDiffusion3Pipeline



model_id = "sd3-reality-mix"
prompt = "ethnographic photography of teddy bear at a picnic holding a sign that says SOON, sitting next to a red sphere which is inside a capsule"
negative_prompt = "malformed, disgusting, overexposed, washed-out"

pipeline = DiffusionPipeline.from_pretrained(model_id)
pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu')
image = pipeline(
    prompt=prompt,
    negative_prompt='blurry, cropped, ugly',
    num_inference_steps=30,
    generator=torch.Generator(device='cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu').manual_seed(1641421826),
    width=1152,
    height=768,
    guidance_scale=5.5,
    guidance_rescale=0.0,
).images[0]
image.save("output.png", format="PNG")