List your company

Competition Mathematical Reasoning on MATH-500 (Accuracy)

66.5AccuracyPEAR
11.3825.694054.31May 26, 2026

Evaluation results

MethodLinks
PEARBackbone=Qwen-3-30B-A3B2026.0566.5
ChainBackbone=Qwen-3-30B-A3B2026.0562
RingBackbone=Qwen-3-30B-A3B2026.0561
RandomBackbone=Qwen-3-30B-A3B2026.0558.5
StarBackbone=Qwen-3-30B-A3B2026.0556
CliqueBackbone=Qwen-3-30B-A3B2026.0554
PEARBackbone=GPT-5.4-nano2026.0552.5
PEARBackbone=Claude-Haiku-4.52026.0549.5
PEARBackbone=Qwen-2.5-14B2026.0548.5
RandomBackbone=GPT-5.4-nano2026.0546
CliqueBackbone=GPT-5.4-nano2026.0545.5
StarBackbone=GPT-5.4-nano2026.0544.5
CoT-SCBackbone=Qwen-3-30B-A3B2026.0544
RingBackbone=GPT-5.4-nano2026.0544
ChainBackbone=GPT-5.4-nano2026.0543.5
CliqueBackbone=Claude-Haiku-4.52026.0542.5
ChainBackbone=Qwen-2.5-14B2026.0542
RandomBackbone=Claude-Haiku-4.52026.0542
RingBackbone=Claude-Haiku-4.52026.0541.5
RingBackbone=Qwen-2.5-14B2026.0541
StarBackbone=Claude-Haiku-4.52026.0541
RandomBackbone=Qwen-2.5-14B2026.0540.5
ChainBackbone=Claude-Haiku-4.52026.0540.5
StarBackbone=Qwen-2.5-14B2026.0539.5
CliqueBackbone=Qwen-2.5-14B2026.0538.5
CoT-SCBackbone=Claude-Haiku-4.52026.0535.5
CoTBackbone=Qwen-3-30B-A3B2026.0534.5
CoT-SCBackbone=GPT-5.4-nano2026.0534
PEARBackbone=Gemma-3-12B2026.0533.5
CoTBackbone=GPT-5.4-nano2026.0531.5
RandomBackbone=Gemma-3-12B2026.0530
ChainBackbone=Gemma-3-12B2026.0529.5
RingBackbone=Gemma-3-12B2026.0529
StarBackbone=Gemma-3-12B2026.0528.5
CoTBackbone=Claude-Haiku-4.52026.0528.5
CliqueBackbone=Gemma-3-12B2026.0527.5
CoT-SCBackbone=Qwen-2.5-14B2026.0527.5
PEARBackbone=Llama-3.1-8B2026.0525.5
CoTBackbone=Qwen-2.5-14B2026.0525.5
ChainBackbone=Llama-3.1-8B2026.0522.5
RandomBackbone=Llama-3.1-8B2026.0522
RingBackbone=Llama-3.1-8B2026.0521.5
CoT-SCBackbone=Gemma-3-12B2026.0520.5
CliqueBackbone=Llama-3.1-8B2026.0520.5
StarBackbone=Llama-3.1-8B2026.0519.5
CoTBackbone=Gemma-3-12B2026.0518.5
CoT-SCBackbone=Llama-3.1-8B2026.0516.5
CoTBackbone=Llama-3.1-8B2026.0513.5