alpaca-first-llama-community-finetune-march-2023
IN premise — summaries/2026/08/24/wiki-LLaMA-chunk-3.md
Created 2026-08-24T17:11:15+00:00
Alpaca (Stanford, March 2023) was the first widely-known community fine-tune of LLaMA-7B (instruction-tuned), pioneering the open fine-tuning ecosystem, and was later taken offline over cost and hallucination concerns.
Summary
Alpaca proved that a small group of researchers could take a base language model and turn it into a practical, instruction-following assistant without a major company behind them, effectively kicking off the wave of open community fine-tunes that followed. Its later removal over cost and reliability issues also sets a precedent that even foundational open projects can be pulled from public use when practical concerns outweigh their community value.