4 ms·
That and because the embedding models are much easier to improve with at scale usage (hence why everyone has a deep search/research/RAG tool built into their AI
by htrp 2y ago
That and because the embedding models are much easier to improve with at scale usage (hence why everyone has a deep search/research/RAG tool built into their AI web app now).