ВАЖНОЕПосты X2026-10-12 17:51 We're launching our leaderboard for FrontierCode 1.1, a private evaluation from Cognition that tests models on completing coding tasks in a way that w... ИСТОЧНИКИ Artificial Analysis2026-10-12 17:51