Vending-Bench Arena

Andon Labs reported on 13 September that OpenAI's GPT-6 Astra averaged a final balance of $15,515 across six runs of the year-long Vending-Bench 2 simulation, nearly tripling Claude Fable 5.1's $5,422…

A study of 2,583 inter-agent emails from 20 one-year simulation runs of Vending-Bench Arena, a competitive vending environment spanning 13 frontier LLMs, found that 12.6% of emails contained misaligne…