FLUX training for 8GB VRAM?

I've tried the options for 12G, 16G, and 20G VRAM options here: https://github.com/kohya-ss/sd-scripts/tree/sd3?tab=readme-ov-file#flux1-lora-training and confirm they all work.

But is it possible to do 8GB VRAM? Is there a specific combination of configs we can try to get it to work with 8GB VRAM?

Start with the FLUX.1 LoRA training section in the linked README and compare the documented 12G, 16G, and 20G VRAM options. Determine whether an 8GB configuration is supported, and document a confirmed configuration or the limitation and relevant requirements.

Written by the indexing model from the issue text.