Argo-Bench Puts Data Agents on a 7.5B-Row Enterprise Warehouse, Top Model Clears Just 34.8% of Tasks
Argo-Bench, released October 1, scores data agents across 210 analytics tasks on a simulated New York food-delivery platform — 81M orders, 235 ...
KaliBench, accepted at NeurIPS 2026's Evaluations and Datasets Track, scores language models on translating analyst instructions into exact Kali ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results