Track 2 Inactive?

Orak Game Agent Challenge
#1

Hi, does nobody like to use large models?

I am a little surprised to see only a couple teams on track 2. Are there still teams waiting to submit initial results (like myself)?

I understand smaller models are usually as capable, but I prefer using API for development anyway to keep local computing lightweight. I’m just worried that prize reallocation will undermine my efforts in track 2 if there is a distinct lack of participation.

Thanks and please let me know thoughts!

#2

i think that maybe people want to try agents on small models which are cheaper to use and then roll out on larger ones if scores will be promising enough, so that to expect to claim credits if goes beyond threshold.