Continual-ARC

ReplayARC-AGI-2 · Fresh · GPT-6 Luna, max

Stateless floor

The model alone, with nothing kept between instances.

Held-out tasks, budget used
27.3 solved 88%
Model spend
$1.66 128 calls
Instances
25 8.9 hours

Budget used is the share of the cap of 16 that an instance used, averaged over the instances: a wrong answer and a demonstration each cost 1, so lower is better.

Loading the run