List your company

Competition Mathematics on AIME 2025 (Acc@t1, Delta Metrics)

80Acc@t1SELF-THOUGHT
3.736823.535943.33563.1341Jan 31, 2026

Evaluation results

MethodLinks
SELF-THOUGHTModel=O3-MINI2026.01806.670
REFLEXModel=O3-MINI2026.0176.673.330
INITIAL (t = 0)Model=O3-MINI2026.0173.33——
SELF-REFINEModel=O3-MINI2026.0173.336.676.67
SELF-THOUGHTModel=DEEPSEEK-R12026.0173.3313.333.33
SELF-REFINEModel=DEEPSEEK-R12026.01702013.33
SELF-TICKModel=O3-MINI2026.0166.671016.67
REFLEXIONModel=O3-MINI2026.0166.673.3310
INITIAL (t = 0)Model=DEEPSEEK-R12026.0163.33——
REFLEXModel=DEEPSEEK-R12026.0163.336.676.67
REFLEXIONModel=DEEPSEEK-R12026.01606.6710
SELF-TICKModel=DEEPSEEK-R12026.0153.336.6716.67
SELF-THOUGHTModel=GPT-4O-MINI2026.012013.330
SELF-REFINEModel=GPT-4O-MINI2026.0116.67100
SELF-THOUGHTModel=GPT-4O2026.0116.67103.33
SELF-TICKModel=GPT-4O-MINI2026.0113.336.670
REFLEXModel=GPT-4O-MINI2026.01103.330
INITIAL (t = 0)Model=GPT-4O2026.0110——
SELF-REFINEModel=GPT-4O2026.01106.676.67
SELF-TICKModel=GPT-4O2026.01106.676.67
REFLEXIONModel=GPT-4O2026.01106.676.67
INITIAL (t = 0)Model=GPT-4O-MINI2026.016.67——
REFLEXIONModel=GPT-4O-MINI2026.016.673.333.33
REFLEXModel=GPT-4O2026.016.6703.33