A highly memory-efficient layer-selective fine-tuning script for Stable Diffusion XL (SDXL) that trains only your chosen UNet layers while freezing the rest, allowing full-quality fine-tuning on just 12 GB VRAM.
flux flow gui ai velocity fine-tuning ai-training sd3 stable-diffusion velocity-prediction rectified-flow sdxl ai-finetune noobai ai-fine-tuning sdxl-training vpred sdxl-vpred eq-vae
-
Updated
Feb 24, 2026 - Python