List your company

Mathematics on AIME 2024 (Accuracy, Throughput)

80AccuracyReasonCache
2841.55568.5Jul 29, 2025

Evaluation results

MethodLinks
ReasonCacheModel=Phi 4 reasoning plus2025.0780245.21
DenseModel=Phi 4 reasoning plus2025.0780172.77
DenseModel=QwQ 32B2025.0776.647.85
DenseModel=DeepSeek R1 Distill Qwen 32B2025.0773.371.59
ReasonCacheModel=QwQ 32B2025.0773.385.3
ReasonCacheModel=DeepSeek R1 Distill Qwen 32B2025.0770123.4
RandomModel=DeepSeek R1 Distill Qwen 32B2025.0736.7120.8
RandomModel=Phi 4 reasoning plus2025.0733.3243.2
RandomModel=QwQ 32B2025.073085.7