Continual-ARC

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

Demo-cache floor

The model alone, shown every demonstration it bought for the task again for free.

Stream, budget used
27.9 solved 76%
Epilogue, budget used
41.3 solved 64%
Model spend
$7.68 1,229 calls
Instances
225 36.8 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.

Effort
Order

Loading the run