24
OLMo – a fully open LLM outperforming GPT 4o mini
(allenai.org)
Fully open is extremely good, it uses entirely open datasets and has all training and inference code. We need more things like that.
Just checked this out on HF and this shit isn't even on the top 100 in any leaderboard. What gives?
Posts from the RSS Feed of HackerNews.
The feed sometimes contains ads and posts that have been removed by the mod team at HN.