List your company
ResearchFormerly WizWand

Mathematical Reasoning on GaoKao (test)

79.74AccuracyDiPO
14.490431.430248.3765.3098Sep 18, 2024Dec 10, 2024Mar 3, 2025May 25, 2025Aug 16, 2025Nov 7, 2025Jan 29, 2026
Updated 8mo ago

Evaluation results

MethodLinks
DiPOModel=Qwen3-4B, Temperature=Constant, Maximum context length=8K tokens2026.0179.741,490.859.3
Qwen2.5-Math-72B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=rm@82024.0976.5——
BaselineModel=Qwen3-4B, Temperature=Constant, Maximum context length=8K tokens2026.0175.583,208.140.2
Qwen2.5-Math-72B-InstructEvaluation Protocol=TOOL-INTEGRATED REASONING, Decoding Strategy=rm@82024.0975.4——
Qwen2.5-Math-72B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=maj@82024.0975——
DiPOModel=Qwen3-8B*, Temperature=Constant, Maximum context length=8K tokens2026.0174.551,575.259.5
Qwen2.5-Math-7B-InstructEvaluation Protocol=TOOL-INTEGRATED REASONING, Decoding Strategy=rm@82024.0972.9——
Qwen2.5-Math-7B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=rm@82024.0972.2——
Qwen2.5-Math-72B-InstructEvaluation Protocol=TOOL-INTEGRATED REASONING, Decoding Strategy=maj@82024.0972——
BaselineModel=Qwen3-8B*, Temperature=Constant, Maximum context length=8K tokens2026.0171.433,976.427.9
Qwen2.5-Math-7B-InstructEvaluation Protocol=TOOL-INTEGRATED REASONING, Decoding Strategy=maj@82024.0970.8——
Qwen2.5-Math-72B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0968.6——
Qwen2.5-Math-72B-InstructEvaluation Protocol=TOOL-INTEGRATED REASONING, Decoding Strategy=Greedy2024.0968.5——
Qwen2.5-Math-1.5B-InstructEvaluation Protocol=TOOL-INTEGRATED REASONING, Decoding Strategy=maj@82024.0968.3——
Qwen2.5-Math-7B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=maj@82024.0968.1——
Qwen2-Math-72B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=rm@82024.0967.7——
GPT-4oCode Integration=No2025.0267.5——
Qwen2.5-Math-1.5B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=rm@82024.0967.5——
Qwen2.5-Math-1.5B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=maj@82024.0966.4——
Qwen2.5-Math-7B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0966.3——
Qwen2.5-Math-1.5B-InstructEvaluation Protocol=TOOL-INTEGRATED REASONING, Decoding Strategy=rm@82024.0964.1——
Qwen2.5-Math-7B-InstructEvaluation Protocol=TOOL-INTEGRATED REASONING, Decoding Strategy=Greedy2024.0962.9——
Qwen2-Math-7B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=rm@82024.0962.7——
Qwen2.5-Math-1.5B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0962.4——
OpenAI-o1-previewCode Integration=No2025.0262.1——
Qwen2-Math-72B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=maj@82024.0961.7——
Qwen2-Math-72B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0959.8——
Qwen2.5-Math-1.5B-InstructEvaluation Protocol=TOOL-INTEGRATED REASONING, Decoding Strategy=Greedy2024.0959.6——
Qwen2-Math-7B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=maj@82024.0959.5——
Qwen2-Math-1.5B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=rm@82024.0958.2——
Qwen2-72B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0954.6——
AutoCode4Math-Qwen2.5Code Integration=Autonomous2025.0251.69——
DeepSeek-Coder-V2-Lite-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0951.1——
AutoCode4Math-DeepSeekCode Integration=Autonomous2025.0250.53——
AutoCode4Math-Qwen2Code Integration=Autonomous2025.0250.13——
Qwen2-Math-1.5B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=maj@82024.0950.1——
NuminaMath-72BCode Integration=Yes2025.0249.4——
Qwen2-Math-7B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0949——
NuminaMath-7B-CoTCode Integration=No2025.0248.83——
NuminaMath-72B-CoTEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0947.9——
Qwen2-Math-1.5B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0946.5——
Mathstral-7BCode Integration=No2025.0246——
Dart-Math-DeepSeek-7BCode Integration=No2025.0245.45——
Qwen-2.5-Base-7BCode Integration=No2025.0245.45——
DeepseekMath-Instruct-7BCode Integration=Yes2025.0244.68——
Qwen2Math-Base-7BCode Integration=No2025.0243.37——
GPT-4o-2024-08-06Evaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0942.6——
Llama-3.1-70B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0941.7——
Internlm2-math-plus-mixtral8x7BEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0937.3——
NuminaMath-7B-CoTEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0936.4——
Internlm2-math-plus-20BEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0936.1——
Qwen2-7B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0935.1——
Dart-Math-Llama3-8BCode Integration=No2025.0234.8——
Internlm2-math-plus-7BEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0934.5——
DeepSeekMath-7B-RLEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0933.6——
TORA-70BCode Integration=Yes2025.0231.7——
Mathstral-7B-v0.1Evaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0931.6——
Llama-3.1-8B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0930.4——
Mammoth-70BCode Integration=Yes2025.0225.2——
Mammoth-Mistral-7BCode Integration=Yes2025.0222.08——
Qwen2-1.5B-InstructEvaluation Protocol=CHAIN-OF-THOUGHT, Decoding Strategy=Greedy2024.0917——