Too many books, too little time...
The AI-Toolkit settings are as follows:
Optimizer: AdamW8bit
Mix Precision: bf16
Save Precision: float16
Learning rate: 1e-4
Unet LR: 1e-4
Linear: 16
Linear_alpha: 16
Resolution: [ 512, 768, 1024 ]
kwBook_FLUX_v1.safetensors
Same model published on other platforms. May have additional downloads or version variants.
