안돈 랩스

소개

Andon Labs is an AI evaluation startup best known for Vending-Bench, a long-horizon benchmark that measures whether language-model agents can profitably and honestly operate a simulated vending-machine business over a full year of simulated time. Its Vending-Bench Arena variant puts multiple models into the same market so they can email, trade, and compete with one another, surfacing behaviours such as price-fixing and prepayment fraud that shorter agentic benchmarks miss.

SNS

검색 트렌드