List your company
ResearchFormerly WizWand

Mathematical Reasoning on AIME 2023

32,881.13Input TokensBoT
-1,028.41327,775.025916,578.46525,381.9041Sep 26, 2025
Updated 7mo ago

Evaluation results

MethodLinks
BoTLLM=Qwen3-8B2025.0932,881.1327,089.730.94
BoTLLM=Qwen3-4B2025.0925,746.7324,447.80.5
BoTLLM=Qwen3-1.7B2025.0917,598.5516,058.920.33
BoTLLM=Qwen3-0.6B2025.098,234.6730,557.50.59
CoT-SCLLM=Qwen3-4B2025.09920.6741,082.130.78
CoT-SCLLM=Qwen3-8B2025.09920.6742,651.131.35
CoT-SCLLM=Qwen3-14B2025.09920.6738,352.22.42
CoT-SCLLM=Qwen3-1.7B2025.09919.1346,885.670.89
CoT-SCLLM=Qwen3-0.6B2025.09906.1343,614.270.83
RAGLLM=Qwen3-0.6B2025.09399.3310,124.530.19
RAGLLM=Qwen3-1.7B2025.09399.3312,881.470.24
RAGLLM=Qwen3-4B2025.09399.3311,270.930.21
RAGLLM=Qwen3-8B2025.09399.3311,378.130.36
RoT + TILLM=Qwen3-0.6B2025.09345.84,932.070.09
RoT + TILLM=Qwen3-1.7B2025.09345.88,222.670.16
RoT + TILLM=Qwen3-4B2025.09345.89,075.20.17
RoT + TILLM=Qwen3-8B2025.09345.810,933.270.35
RoT + TILLM=Qwen3-14B2025.09345.89,421.530.6
CoTLLM=Qwen3-0.6B2025.09285.69,719.330.18
CoTLLM=Qwen3-1.7B2025.09285.612,826.270.24
CoTLLM=Qwen3-4B2025.09285.611,202.730.21
CoTLLM=Qwen3-8B2025.09285.612,4850.39
CoTLLM=Qwen3-14B2025.09285.69,165.670.58
RAGLLM=Qwen3-14B2025.09277.139,633.730.61
RoTLLM=Qwen3-0.6B2025.09275.811,718.20.22
RoTLLM=Qwen3-1.7B2025.09275.812,515.930.24
RoTLLM=Qwen3-4B2025.09275.811,358.80.22
RoTLLM=Qwen3-8B2025.09275.812,555.870.4
RoTLLM=Qwen3-14B2025.09275.89,911.20.63