Vending-Bench 2

Measures: An agent runs a simulated vending-machine business for one simulated year with tools, handling suppliers (some adversarial), inventory, pricing and disruptions over tens of millions of tokens; the score is the bank balance at year end, starting from $500 and averaged over about 5 runs.

Maintainer: Andon Labs
Introduced: 2025
Status: Active
Saturation: Headroom
Typical provenance: Independent
In scoring set: No

Not scored: Open-ended dollar metric with no hard ceiling; Andon Labs' estimate that a good strategy earns about $63k could become the reference ceiling in a later method version.

Paper Official leaderboard

Leaderboard

Show independent results only

Every result we hold, whoever ran it. Only independent and third-party rows count towards a score.

No results to show.