Share your thoughts, 1 month free Claude Pro on usSee more

Mathematical Reasoning on OlympiadBench Math

99.86Accuracy

HEART

Updated 1mo ago

Evaluation Results

Method	Links
HEART 2025.09		99.86
HEART 2025.09		94.63
HEART 2025.09		94.44
HEART 2025.09		93.63
HEART 2025.09		92.59
Vanilla 2025.09		83.63
Vanilla 2025.09		77.67
AM-Thinking (math) 2025.12		77.5
MiroMind-M1-SFT 2025.12		77
Vanilla 2025.09		76.93
ODA-Math-460k 2025.12		76.3
Vanilla 2025.09		75.44
OmniThought-0528 2025.12		74.9
Vanilla 2025.09		74.41
AM-Thinking (math) 2025.12		74.2
HEART 2025.09		72.78
SYNTHETIC-2-SFT 2025.12		71.5
ODA-Math-460k 2025.12		70.9
Light-R1-SFT 2025.12		69.7
OpenThoughts3 2025.12		68.8
OmniThought-0528 2025.12		68.1
SYNTHETIC-2-SFT 2025.12		67.4
MiroMind-M1-SFT 2025.12		66.3
Fast-Math-R-SFT 2025.12		61
DeepMath-103K 2025.12		60.2
Light-R1-SFT 2025.12		60.2
HEART 2025.09		58.63
MegaScience (math) 2025.12		57.6
Vanilla 2025.09		52.33
Fast-Math-R1-SFT 2025.12		50.3
INFUSER 2026.06		50.24
AZR 2026.06		47.92
Qwen3-8B-Base 2025.12		47.2
PivotTrace 2026.06		47
CoE 2026.06		46.8
Random 2026.06		46.7
R-Few† 2026.06		46.4
Entropy 2026.06		46.3
Self-Certainty 2026.06		45.8
CoT-Kinetics 2026.06		45.7
Consistency 2026.06		45.2
R-Zero 2026.06		45.1
MegaScience (math) 2025.12		44.5
SPICE† 2026.06		42.5
Vanilla 2025.09		40.7
Base 2026.06		40.36
Qwen2.5-7B-Base 2025.12		35.9
LIMO 2025.12		34.9
LIMO 2025.12		31.3
OpenMathInstruct-2 2025.12		30.7
MathFusion-DSMath-7B 2025.03		25.5
MathFusion-DSMath-7B 2025.03		23.3
DeepSeekMath-7B-DART-Math 2025.03		21.7
DART-Math-DSMath-7B 2024.06		21.7
MathFusion-DSMath-7B 2025.03		21.6
DART-Math-DSMath-7B 2024.06		21.3
DeepSeekMath-7B-DART-Math† 2025.03		21
MathFusion-DSMath-7B 2025.03		19.3
DeepSeekMath-7B-RFT 2025.03		19.1
MathFusion-DSMath-7B 2025.03		19
DeepSeekMath-7B-RL 2024.06		18.7
MathFusion-Llama3-8B 2025.03		17.2
Mistral-7B-DART-Math 2025.03		14.7
Llama3-8B-DART-Math 2025.03		14.5
DeepSeekMath-7B-Instruct 2025.03		14.2
MathFusion-Mistral-7B 2025.03		13.6
DeepSeekMath-7B-MMIQC 2025.03		13
Llama3-8B-DART-Math† 2025.03		12.9
MathFusion-Llama3-8B 2025.03		12.6
MathFusion-Llama3-8B 2025.03		11.9
MathFusion-Llama3-8B 2025.03		11.9
MathFusion-Mistral-7B 2025.03		11
DeepSeekMath-7B-RefAug 2025.03		10.5
DeepSeekMath-7B-MMIQC† 2025.03		10.4
DeepSeekMath-7B-RefAug 2025.03		10.1
Llama3-8B-MMIQC 2025.03		9.6
DeepSeekMath-7B-MetaMath 2025.03		9.5
Mistral-7B-MMIQC 2025.03		9.4
MathFusion-Mistral-7B 2025.03		9.3
Llama3-8B-RFT 2025.03		9.3
Mistral-7B-RFT 2025.03		8.7
Mistral-7B-DART-Math† 2025.03		8.7
Mistral-7B-WizardMath-V1.1 2025.03		7.7
MathFusion-Mistral-7B 2025.03		7.3
Mistral-7B-MetaMath 2025.03		5.9
DeepSeekMath-7B-Standard 2025.03		5.6
Llama3-8B-MetaMath 2025.03		5.5
Llama3-8B-RefAug 2025.03		5.5
Llama3-8B-MetaMath† 2025.03		5.3
Llama3-8B-MMIQC† 2025.03		5.2
Mistral-7B-MetaMath† 2025.03		5
Mistral-7B-MMIQC† 2025.03		5
Llama3-8B-Standard 2025.03		4.7
Llama3-8B-RefAug 2025.03		4.7
Mistral-7B-RefAug 2025.03		3.9
Mistral-7B-RefAug 2025.03		3.1
Mistral-7B-Standard 2025.03		2.2