oobabooga's textgen webui has a tab for fine tuning now. You only need a single consumer GPU to fine tune up to 33B parameter models at a rate of about 200 epochs per hour, per GPU.

There are also one-click finetuning projects which run on free Google Colab GPUs like https://github.com/lxe/simple-llama-finetuner

It's easy and not complex at all.