README.md

metadata

license: creativeml-openrail-m
base_model: terminusresearch/pixart-900m-1024
tags:
  - stable-diffusion
  - stable-diffusion-diffusers
  - text-to-image
  - diffusers
  - full
inference: true

pixart-900m-1024-ft-large

This is a full rank finetune derived from terminusresearch/pixart-900m-1024.

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: 8.5
• CFG Rescale: 0.0
• Steps: 30
• Sampler: euler
• Seed: 42
• Resolutions: 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:

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

Training settings

• Training epochs: 1
• Training steps: 6500
• Learning rate: 1e-06
• Effective batch size: 384
  • Micro-batch size: 24
  • 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: ~559104
• Total number of aspect buckets: 1
• Resolution: 1.0 megapixels
• Cropped: True
• Crop style: random
• Crop aspect: square

dalle3

• Repeats: 0
• Total number of images: ~972672
• Total number of aspect buckets: 1
• Resolution: 1.0 megapixels
• Cropped: True
• Crop style: center
• Crop aspect: square

nijijourney-v6-520k-raw

• Repeats: 0
• Total number of images: ~415872
• Total number of aspect buckets: 1
• Resolution: 1.0 megapixels
• Cropped: True
• Crop style: center
• Crop aspect: square

midjourney-v6-520k-raw

• Repeats: 0
• Total number of images: ~390912
• Total number of aspect buckets: 1
• Resolution: 1.0 megapixels
• Cropped: True
• Crop style: center
• Crop aspect: square

Inference

import torch
from diffusers import DiffusionPipeline



model_id = "pixart-900m-1024-ft-large"
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',
    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=8.5,
    guidance_rescale=0.0,
).images[0]
image.save("output.png", format="PNG")