ResearchBenchmarksMathematical Reasoning on Gaokao2023 (test)Mathematical Reasoning on Gaokao2023 (test)Follow57.4AccuracyAgentic-Verifier-Qwen3-4B24.43232.99141.5550.109Apr 17, 2026Evaluation resultsMethodMethodLinksAccuracyAgentic-Verifier-Qwen3-4BBoN=@128BoN=@1282026.0457.4Agentic-Verifier-Qwen3-4BBoN=@64BoN=@642026.0455.1Agentic-Verifier-Qwen3-4BBoN=@32BoN=@322026.0454.5Qwen2.5-Math-PRM-7BBoN=@128BoN=@1282026.0454.3Qwen3-4B-ThinkBoN=@32BoN=@322026.0452.2Qwen3-4B-ThinkBoN=@128BoN=@1282026.0451.9Qwen3-4B-ThinkBoN=@64BoN=@642026.0451.4Qwen2.5-Math-PRM-7BBoN=@64BoN=@642026.0451.4Qwen2.5-Math-PRM-7BBoN=@32BoN=@322026.0450.9DS-Distill-14BBoN=@64BoN=@642026.0449.1DS-Distill-14BBoN=@128BoN=@1282026.0448.6DS-Distill-14BBoN=@32BoN=@322026.0448.3Agentic-Verifier-Llama3-8BBoN=@64BoN=@642026.0447.3Skywork-PRMBoN=@64BoN=@642026.0447Agentic-Verifier-Llama3-8BBoN=@128BoN=@1282026.0447Skywork-PRMBoN=@128BoN=@1282026.0446.8Skywork-PRMBoN=@32BoN=@322026.0446Agentic-Verifier-Llama3-8BBoN=@32BoN=@322026.0445.5INF-ORM-Llama3.1-70BBoN=@128BoN=@1282026.0444.4InternLM2-20B-RMBoN=@128BoN=@1282026.0443.9Math-Shepherd-Mistral-7B-PRMBoN=@32BoN=@322026.0443.1InternLM2-20B-RMBoN=@64BoN=@642026.0442.9INF-ORM-Llama3.1-70BBoN=@64BoN=@642026.0442.6Qwen2.5-7B-InstructBoN=@64BoN=@642026.0442.3Skywork-V2-Llama-8BBoN=@32BoN=@322026.0441.6InternLM2-20B-RMBoN=@32BoN=@322026.0441.6Math-Shepherd-Mistral-7B-PRMBoN=@64BoN=@642026.0441.4Qwen2.5-7B-InstructBoN=@128BoN=@1282026.0441.3Qwen2.5-7B-InstructBoN=@32BoN=@322026.0440.8INF-ORM-Llama3.1-70BBoN=@32BoN=@322026.0440.8Starling-RM-34BBoN=@32BoN=@322026.0440.5Mistral-Small-24B-InstructBoN=@128BoN=@1282026.0440Skywork-V2-Llama-8BBoN=@128BoN=@1282026.0439.7Starling-RM-34BBoN=@64BoN=@642026.0439.7Math-Shepherd-Mistral-7B-PRMBoN=@128BoN=@1282026.0439.3Mistral-Small-24B-InstructBoN=@32BoN=@322026.0439Starling-RM-34BBoN=@128BoN=@1282026.0439Mistral-Small-24B-InstructBoN=@64BoN=@642026.0438.4Skywork-V2-Llama-8BBoN=@64BoN=@642026.0437.4Llama3.1-8B-InstructBoN=@64BoN=@642026.0435.6Llama3.1-8B-InstructBoN=@128BoN=@1282026.0435.6Llama3.1-8B-InstructBoN=@32BoN=@322026.0434.5GRM-Gemma-2BBoN=@32BoN=@322026.0433.5GRM-Gemma-2BBoN=@128BoN=@1282026.0433.2EurusPRM-Stage1BoN=@32BoN=@322026.0433.2GRM-Gemma-2BBoN=@64BoN=@642026.0431.9EurusPRM-Stage1BoN=@64BoN=@642026.0429.1EurusPRM-Stage2BoN=@32BoN=@322026.0429.1EurusPRM-Stage1BoN=@128BoN=@1282026.0426.8EurusPRM-Stage2BoN=@64BoN=@642026.0426.5EurusPRM-Stage2BoN=@128BoN=@1282026.0425.7