external benchmark
Vending-Bench 2 leaderboard
Every model the catalog carries a published Vending-Bench 2 value for, ranked by that value.
Categoryexternal
MeasureArithmetic mean final bank balance in USD
TasksOne year of simulated autonomous business operation per run
DifficultyLong-horizon agent coherence and business operation
Published byVending-Bench 2
Tests long-horizon coherence by asking agents to operate a simulated vending-machine business for one year.
No model in the catalog has a published Vending-Bench 2 score.
The benchmark is defined by Vending-Bench 2, but the catalog carries no value for it yet. An absent value is shown as absent here rather than as a zero.