Local 1M Context Inference at 15 tokens/s and ~100% "Needle In a Haystack" (old.reddit.com) 1 points by quxinxin 2y ago ↗ HN
0 comments
[ 3.2 ms ] story [ 13.6 ms ] threadNo comments yet.