3 ms·
current LLMs are too large or efficient enough for local use needs more breakthroughs on both chip side and model side
by youarenotyu 4mo ago
current LLMs are too large or efficient enough for local use
needs more breakthroughs on both chip side and model side