From GPT-4 to Mistral 7B, there is a 300x range in the cost of LLM inference (twitter.com) 2 points by Gcam 2y ago ↗ HN
0 comments
[ 4.3 ms ] story [ 13.5 ms ] threadNo comments yet.