List your company

Proof Verification on 21-proof (Diagnostic Set)

9Correct Valid ProofsCodex 5.4 Sentence-by-Sentence
7.26327.71418.1658.6159Jun 9, 2026

Evaluation results

MethodLinks
Codex 5.4 Sentence-by-SentenceEvaluation Strategy=Sentence-by-Sentence, Trials=avg@32026.0692910.8571
Global GPT-5.4Evaluation Strategy=Global, Trials=avg@32026.068.332.677.672.330.7619
AgentEvaluation Strategy=Constructive Verification, Trials=avg@32026.06831000.8571
Global Codex-5.4Evaluation Strategy=Global, Trials=avg@32026.067.333.67820.73