SDXL Training: AdamW Batch Size 30 vs Adafactor Batch Size 1 Comparison
SDXL training test on an RTX A6000: AdamW at batch size 30 vs the SECourses Adafactor config at batch size 1, which gave far better resemblance.

Highlights
AdamW used 47 GB VRAM and was cooked after 450 epochs while still behind in resemblance. Both runs used 15 training and 15 regularization images on Massed Compute with similar total time, and the full 321 MB grid of 20 special prompts is on Hugging Face.
Video tutorial
Train Stable Diffusion on 15 Photos of Your Face: Full SDXL Fine-Tuning With OneTrainer (2024) · chapters and summary
Related

Kohya vs OneTrainer for FLUX Dev Fine-Tuning / DreamBooth: Comparison
Kohya and OneTrainer compared for FLUX Dev fine-tuning: OneTrainer with EMA and a copied Kohya config gave weaker results than the best Kohya config.

FLUX 2 vs FLUX SRPO & New Kohya SS GUI Premium FLUX Training App
FLUX 2 Pro compared with a trained FLUX SRPO model, plus the new Kohya trainer V35 that trains FLUX on GPUs with as little as 6 GB VRAM.

FLUX LoRA vs Fine Tuning / DreamBooth Comparison: Batch Size 1 vs 7
FLUX LoRA and Fine Tuning / DreamBooth compared with 15 and 256 training images at batch size 1 and 7, for both realism and stylization.