This llama model was trained 2x faster with Unsloth

This is a finetune of Llama-3.2:1B, meant to take on the personality of my OC - Aoi, a shy, professional girl with some laser focus when she needs to. this was really meant to be a fun experiment with finetuning, so if you encounter anything that is of note, feel free to notify me.
DEPRECATED: I did not bother to quantize it, so if you want to quantize it, feel free to go ahead. as I don't know how this all works yet, if you want to try out the model for yourself, you will have to create it using the GGUF and the Modelfile. To do that, download the .gguf and the Modelfile (make sure they are in the same place) and then open terminal, cd to the directory that has those two files saved in, and then run the create model command for your specified platform. in ollama it is ollama create aoi -f Modelfile. and now you can run it!
UPDATE 1: uploaded the Q4_K_M quant as a .gguf file. still no idea how to have it show on the right side (if it's supposed to at all), so you'll have to manually download it or pull it using a something that runs .gguf (llama.cpp or Ollama, etc.)