1Show HN: Stirrup – A lightweight and customizable foundation for building agents (opens in new tab)(github.com)2Gcam3mo ago0
2MicroEvals – Easily run vibe checks against models (opens in new tab)(artificialanalysis.ai)3Gcam9mo ago0
3From GPT-4 to Mistral 7B, there is a 300x range in the cost of LLM inference (opens in new tab)(twitter.com)2Gcam2y ago0
4Show HN: LLM Benchmarks Leaderboard with 60 model and API host combinations (opens in new tab)(artificialanalysis.ai)3Gcam2y ago1
5Mistral API reduces time to first token by 10x (only place for Mistral Medium) (opens in new tab)(twitter.com)4Gcam2y ago0
6240 Tokens/s achieved by Groq's custom chips on Lama 2 Chat (70B) (opens in new tab)(twitter.com)5Gcam2y ago0
7New GPT-4 Turbo (0125 Preview) slightly faster per initial benchmarks (opens in new tab)(twitter.com)2Gcam2y ago0