Cognition announces GPT-6.1 Sol low reasoning tier evaluation results

Cognition announced the evaluation results of GPT-6.1 Sol at the low reasoning tier. It scored 58.1%, with a per-task cost of $0.21, higher than GPT-6 Sol's 50.5% at the same tier. Cognition stated that this is the highest score among models on the leaderboard with a per-task cost below $0.30, and also provided a trial entry point. There are currently no further developments.

AI-generated from public reports

1. Sep 29· 2 reports

   Cognition announces GPT-6.1 Sol evaluation results

   Source:Cognition