List your company

Mathematical Reasoning on AIME 2024 (Accuracy and Token Counts vs Base)

77.9AccuracyTHINKBRAKE
0.31620.45840.660.742Oct 1, 2025Oct 11, 2025Oct 22, 2025Nov 2, 2025Nov 13, 2025Nov 24, 2025Dec 5, 2025

Evaluation results

MethodLinks
THINKBRAKEBase Model=Qwen3-14B2025.1077.9———-7.4
Qwen3-4B-ThinkingStrategy=Baseline2025.1077.1————
THINKBRAKEBase Model=Qwen3-32B2025.1077.1———-6.2
Qwen3-32BStrategy=Baseline2025.1075.8————
THINKBRAKEBase Model=Phi-4-Reasoning2025.1073.3———-6.4
Qwen3-14BStrategy=Baseline2025.1071.7————
BaselineBackbone=QwQ-32B, Temperature=0.02025.1270—9,590.47——
NoWaitBase Model=Qwen3-14B2025.1070———-19.8
Qwen3-4BStrategy=Baseline2025.1068.8————
Phi-4-ReasoningStrategy=Baseline2025.1067.5————
THINKBRAKEBase Model=Qwen3-4B-Thinking2025.1067.2———-33.5
NoWaitBase Model=Qwen3-4B-Thinking2025.1066.7———-30.5
NoWaitBase Model=Qwen3-32B2025.1066.7———-12
DEERBackbone=QwQ-32B, Temperature=0.62025.1266.6711.128,767.23-13.3—
LYNXBackbone=QwQ-32B, Conf=0.97, Temperature=0.02025.1266.67-4.768,440.9-12—
THINKBRAKEBase Model=Qwen3-4B2025.1064.6———-12.7
LYNXBackbone=QwQ-32B, Conf=0.95, Temperature=0.02025.1263.33-9.537,725.73-19.4—
LYNXBackbone=QwQ-32B, Conf=0.90, Temperature=0.02025.1263.33-9.537,308.63-23.8—
DEERBase Model=Qwen3-4B-Thinking2025.1063.3———-34.3
Dynasor-CoTBase Model=Qwen3-4B-Thinking2025.1063.3———-28.8
NoWaitBase Model=Phi-4-Reasoning2025.1063.3———-9.4
BaselineBackbone=QwQ-32B, Temperature=0.62025.1260—10,115.93——
LYNXBackbone=QwQ-32B, Conf=0.97, Temperature=0.62025.126007,343.97-27.4—
LYNXBackbone=QwQ-32B, Conf=0.95, Temperature=0.62025.126006,743.93-33.3—
DEERBackbone=QwQ-32B, Temperature=0.02025.1260-14.299,261.73-3.4—
NoWaitBase Model=Qwen3-4B2025.1056.7———-16.6
ToNBackbone=QwQ-32B, Temperature=0.02025.1253.33-23.819,612.260.2—
ToNBackbone=QwQ-32B, Temperature=0.62025.1250-16.678,111.83-19.8—
LYNXBackbone=QwQ-32B, Conf=0.9, Temperature=0.62025.1250-16.675,940.8-41.3—
LYNXBackbone=QwQ-32B, Conf=0.80, Temperature=0.02025.1250-28.576,396.4-33.3—
DeepSeek-R1-7BStrategy=Baseline2025.1049.5————
LYNXBackbone=QwQ-32B, Conf=0.8, Temperature=0.62025.1246.67-22.225,262.7-48—
LYNXBackbone=QwQ-32B, Conf=0.70, Temperature=0.02025.1246.67-33.335,691.37-40.7—
THINKBRAKEBase Model=DeepSeek-R1-7B2025.1045.2———-25
NoWaitBase Model=DeepSeek-R1-7B2025.1040———-32.4
THINKBRAKE-pBase Model=Qwen3-4B-Thinking2025.1035———-72.6
LYNXBackbone=QwQ-32B, Conf=0.7, Temperature=0.62025.1233.33-44.454,248.9-58—
DEERBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Temperature=0.62025.123079.968,303.06-31.9—
BaselineBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Temperature=0.02025.1223.33—13,427.43——
ToNBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Temperature=0.02025.1223.33011,928.16-11.2—
LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.97, Temperature=0.02025.1223.33013,322.6-0.8—
LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.95, Temperature=0.02025.1223.33013,004.03-3.2—
LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.80, Temperature=0.02025.1223.3308,501.83-36.7—
LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.70, Temperature=0.02025.1223.3305,981.83-55.5—
ToNBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Temperature=0.62025.122019.9810,408.63-14.6—
DEERBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Temperature=0.02025.1220-14.279,154.13-31.8—
LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.90, Temperature=0.02025.1220-14.2711,635.8-13.3—
BaselineBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Temperature=0.62025.1216.67—12,189——
LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.97, Temperature=0.62025.1216.67010,817.13-11.3—
LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.95, Temperature=0.62025.1216.6709,963-18.3—
LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.9, Temperature=0.62025.1213.33-20.047,814.43-35.9—
LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.8, Temperature=0.62025.1213.33-20.045,503.27-54.9—
LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.7, Temperature=0.62025.1210-40.013,900.03-68—
ThinkLessBase Model=Qwen3-4B-Thinking2025.103.3———-100