MAVEN Boosts Agentic Tool‑Calling Accuracy to 71% with Minimal Cost
MAVEN introduces a lightweight verification scaffold that lets large language models better decompose tasks and orchestrate tools across varied environments. In the new MAVEN‑Bench stress tests, it lifts a GPT‑OSS‑120B model’s accuracy from 48% to 71% without extra training, matching proprietary baselines at a tenth of the cost.