List your company

Mathematical Reasoning on AMC 2023 (Acc, Len, CR)

97.5AccuracyASAG
71.578.258591.75Jun 13, 2026

Evaluation results

MethodLinks
ASAGModel=Qwen3-32B2026.0697.54,83062.6
ASAGModel=Qwen3-14B2026.06954,38960.8
VanillaModel=Qwen3-32B2026.06957,711100
ASAGModel=Qwen3-8B2026.0692.54,28553.3
VanillaModel=Qwen3-14B2026.0692.57,214100
ASAGModel=Qwen3-4B2026.06905,23069.6
VanillaModel=Qwen3-4B2026.0687.57,512100
VanillaModel=Qwen3-8B2026.0687.58,045100
ASAGModel=DeepSeek-R1-Distill-Qwen-7B2026.0682.54,31263
ASAGModel=DeepSeek-R1-Distill-Llama-8B2026.06804,08763.1
VanillaModel=DeepSeek-R1-Distill-Qwen-7B2026.0677.56,841100
VanillaModel=DeepSeek-R1-Distill-Llama-8B2026.0672.56,482100