List your company
ResearchFormerly WizWand

Mathematical Reasoning on AIME 2024 (Latency)

29.58Latency (s)RoT + TI
-7.9988245.6581499.315752.9719Sep 26, 2025
Updated 7mo ago

Evaluation results

MethodLinks
RoT + TIModel=Qwen3-0.6B2025.0929.58
RoT + TIModel=Qwen3-1.7B2025.09112.52
RoTModel=Qwen3-0.6B2025.09141.02
RoTModel=Qwen3-1.7B2025.09148.85
CoTModel=Qwen3-1.7B2025.09150.75
RAGModel=Qwen3-0.6B2025.09157.17
RoT + TIModel=Qwen3-4B2025.09158.86
RAGModel=Qwen3-1.7B2025.09165
CoTModel=Qwen3-4B2025.09166.24
CoTModel=Qwen3-0.6B2025.09171.01
RAGModel=Qwen3-4B2025.09172.58
RoTModel=Qwen3-4B2025.09177.89
RoT + TIModel=Qwen3-8B2025.09240.14
CoTModel=Qwen3-8B2025.09245.56
CoTModel=Qwen3-14B2025.09252.21
RoTModel=Qwen3-8B2025.09255.75
RAGModel=Qwen3-8B2025.09267.58
RAGModel=Qwen3-14B2025.09291.2
RoTModel=Qwen3-14B2025.09293.53
RoT + TIModel=Qwen3-14B2025.09295.89
BoTModel=Qwen3-0.6B2025.09337.85
BoTModel=Qwen3-1.7B2025.09345.94
CoT-SCModel=Qwen3-1.7B2025.09401.14
BoTModel=Qwen3-4B2025.09410.22
BoTModel=Qwen3-8B2025.09556.07
CoT-SCModel=Qwen3-0.6B2025.09611.43
CoT-SCModel=Qwen3-4B2025.09692.94
CoT-SCModel=Qwen3-8B2025.09900.96
CoT-SCModel=Qwen3-14B2025.09969.05