Demo-cache floor
The model alone, shown every demonstration it bought for the task again for free.
- Stream, budget used
- 16.5 solved 93%
- Epilogue, budget used
- 33.0 solved 84%
- Model spend
- $12.59 881 calls
- Instances
- 225 59.8 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.