Top New Best Ask Show Jobs

Llama 3.1: Our most capable models to date | Hacker News Reader

Llama 3.1: Our most capable models to date

ai.meta.com

27 points·langitbiru··3 comments

Open articleView on HN

[dupe]

More discussion: https://news.ycombinator.com/item?id=41046540

https://news.ycombinator.com/item?id=41046773

405B is already being served on WhatsApp!

https://ibb.co/kQ2tKX5

MMLU PRO is the benchmark I trust the most. I noticed they are using 5 shots and CoT. Is that true for GPT4 and Sonnet as well?

Reply on news.ycombinator.com