Llama3.2-Instruct

1 run on 1 machine by 1 person.

Run it Get the model

By machineEach machine's best run of this model.

  1. 15/24tasks passed, not a score
    M3 Pro 18 GBApple silicon · llama.cpp · resyn.cloud in the browser · 34.7 tok/s · 15/24 tasks · 5.1k tokens · 1 run by uwu

    Not on the leaderboard: it is a fast run, and only full runs take a place on a computer board.

Hardware leaderboard

Latest runs

  1. 15/24
    M3 Pro 18 GB34.7 tok/s · 5.1k tokens · llama.cpp · uwu · 4 days ago

Where people got itLinks as people gave them.

Who ran it

Follow it

Atom feed of new runs

Comments on this model.

Questions about this model? Ask in the chat.

Comments

Sign in or create an account to comment.