Continual-ARC

ReplayARC-AGI-2 · Sequential · GPT-6 Luna, xhigh

Prime Agent, memory wiped

Everything the harness wrote deleted before every instance.

Stream, budget used
34.7 solved 74%
Epilogue, budget used
36.3 solved 68%
Model spend
$7.52 3,730 calls
Instances
225 21.4 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.

Variant
Effort
Order

Loading the run