Lean Verifier: Four Models, Five Rows
GPT-6.1 Sol, Opus 5.5, Kimi K3 and GLM-5.3 on five lean-verifier rows from easy to extremely hard, with the full trace of every run and how the grader decided it.
GPT-6.1 Sol, Opus 5.5, Kimi K3 and GLM-5.3 on five lean-verifier rows from easy to extremely hard, with the full trace of every run and how the grader decided it.