List your company

Mathematical Theorem Proving on OlympiadBench

11.9Proof-level AccuracyBASE
11.17211.36111.5511.739Jun 14, 2026

Evaluation results

MethodLinks
BASESolver=Gemini 3 Pro, Evaluation Protocol=BASE, Prover=DeepSeek-Prover-V2, K=82026.0611.9
BASESolver=R1-Qwen3-8B, Evaluation Protocol=BASE, Prover=DeepSeek-Prover-V2, K=82026.0611.8
BASESolver=Qwen3-8B, Evaluation Protocol=BASE, Prover=DeepSeek-Prover-V2, K=82026.0611.2