(mistral.ai)

701 points mfiguiere | 1 comments | 21 May 25 14:21 UTC | HN request time: 0.2s | source

Show context

gyudin ◴[21 May 25 17:32 UTC] No.44053907[source]▶

Super weird benchmarks

avereveard ◴[21 May 25 17:36 UTC] No.44053952[source]▶

from what I gather it's finetuned to use OpenHand specifically so shows value on thsoe benchmark that target a whole system as a blackbox (i.e. agent + llm) more than directly target the llm input/outputs

replies(1): >>44057074 #

1. amarcheschi ◴[21 May 25 22:54 UTC] No.44057074[source]▶

>>44053952 #

Yup the 1st comment says this https://www.reddit.com/r/LocalLLaMA/comments/1kryybf/mistral...

↑

Devstral