Locally run an Instruction-Tuned Chat-Style LLM

Summary: A fast ChatGPT-like model can now be run on local devices using a combination of the LLaMA foundation model and an open reproduction of Stanford Alpaca, according to a GitHub repository by developer antimatter15. The model has been fine-tuned to obey instructions and can be downloaded along with weights for ggml-alpaca-7b-14.bin using BitTorrent or IPFS.