terminus-xl-velocity-training

This is a full rank finetuned model derived from ptx0/terminus-xl-velocity-v2.

The main validation prompt used during training was:

a cute anime character named toast holding a sign that says SOON, sitting next to a red square on her left side, and a transparent sphere on her right side

Validation settings

• CFG: 7.5
• CFG Rescale: 0.7
• Steps: 30
• Sampler: euler
• Seed: 42
• Resolutions: 1024x1024,1152x960,896x1152

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

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

Training settings

• Training epochs: 0
• Training steps: 800
• Learning rate: 4e-07
• Effective batch size: 16
  • Micro-batch size: 4
  • Gradient accumulation steps: 4
• Prediction type: v_prediction
• Rescaled betas zero SNR: True
• Optimizer: AdamW, stochastic bf16
• Precision: Pure BF16
• Xformers: Not used

Datasets

training-test

• Repeats: 1
• Total number of images: 192
• Total number of aspect buckets: 17
• Resolution: 1.0 megapixels
• Cropped: True
• Crop style: corner
• Crop aspect: preserve

dalle3

• Repeats: 1
• Total number of images: 57744
• Total number of aspect buckets: 2
• Resolution: 1.0 megapixels
• Cropped: False
• Crop style: None
• Crop aspect: None

photo-concept-bucket

• Repeats: 0
• Total number of images: 70616
• Total number of aspect buckets: 5
• Resolution: 1.0 megapixels
• Cropped: True
• Crop style: random
• Crop aspect: random