News for hackers. Stuff that matters.

A comment on Performance of LLMs on Advent of Code 2024.

Re:Performance of LLMs on Advent of Code 2024 by upghost on Monday December 30, 2024 @06:18PM (#42554711)
After looking at the charts I was like "Whoa, damn, that Jerpint model seems amazing. Where do I get that??" I spent some time trying to find it on Huggingface before I realized...
[ Parent | on HN ]
Re:Performance of LLMs on Advent of Code 2024 by j45 on Monday December 30, 2024 @06:56PM (#42554955)
Me too.

The author could make a model on huggingface routing requests to him. Latency might vary.

[ Parent | on HN ]
Re:Performance of LLMs on Advent of Code 2024 by tbagman on Monday December 30, 2024 @09:28PM (#42555785)
The good news is that training jerprint was probably cheaper than training the latest got models...
[ Parent | on HN ]
Re:Performance of LLMs on Advent of Code 2024 by senordevnyc on Monday December 30, 2024 @06:50PM (#42554919)
lol, I did the same thing
[ Parent | on HN ]
Re:Performance of LLMs on Advent of Code 2024 by foldl2022 on Monday December 30, 2024 @08:25PM (#42555461)
Just open-wights it. We need this. :)
[ Parent | on HN ]