GGUF quantization of https://huggingface.co./Finnish-NLP/llama-7b-finnish-instruct-v0.2 – see details there.
5-bit