Y Hacker News new | ask | show | jobs

user: Gcam created: 2016-07-11 karma: 87 twitter: https://twitter.com/grmcameron submissions: 0 points | 0 comments Show HN: Stirrup – A lightweight and customizable foundation for building agents 2 points | 0 comments 0 points | 0 comments 0 points | 0 comments MicroEvals – Easily run vibe checks against models 3 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments From GPT-4 to Mistral 7B, there is a 300x range in the cost of LLM inference 2 points | 0 comments Show HN: LLM Benchmarks Leaderboard with 60 model and API host combinations 3 points | 1 comments Mistral API reduces time to first token by 10x (only place for Mistral Medium) 4 points | 0 comments 240 Tokens/s achieved by Groq's custom chips on Lama 2 Chat (70B) 5 points | 0 comments 0 points | 0 comments New GPT-4 Turbo (0125 Preview) slightly faster per initial benchmarks 2 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments 0 points | 0 comments