well this was a waste of tokens.
presented without commentary:
Prime Agent is a general-purpose coding harness
On ARC-AGI-3, it scores 95.5%, surpassing the human-expert baseline, but the gain is not benchmark-specific.
We see major improvements across models when compared to their proprietary harnesses: https://t.co/Xa8zZiDtbF
On ARC-AGI-3, it scores 95.5%, surpassing the human-expert baseline, but the gain is not benchmark-specific.
We see major improvements across models when compared to their proprietary harnesses: https://t.co/Xa8zZiDtbF
for repro: I gave it a py kernel just like they did (which we had for about 6mo now btw)
bascially ends up spamming eval as you can imagine:
Missing some Tweet in this thread? You can try to Update
