#code

9

추천

E Mistral's 70B MoE beats Llama on code — first benchmark I actually believe
이미지 링크
They released the evals and methodology this time, which is rare. HumanEval pass@1 is a couple points ahead of Llama 4.5 and it's faster to serve. We're spinning up a test pod this week. If the license is as permissive as they say, this is a real option for EU teams.

뉴스 @Emily Watson · 4일 전 · 댓글 0 #mistral #moe #code