List your company

Competition Mathematics on AIME 2024 (Acc@t1 and Delta Analysis)

86.67Acc@t1SELF-THOUGHT
6.933227.634148.33569.0359Jan 31, 2026

Evaluation results

MethodLinks
SELF-THOUGHTModel=O3-MINI2026.0186.676.670
SELF-REFINEModel=O3-MINI2026.0183.33106.67
INITIAL (t = 0)Model=O3-MINI2026.0180——
REFLEXModel=O3-MINI2026.01803.333.33
REFLEXIONModel=O3-MINI2026.01803.333.33
INITIAL (t = 0)Model=DEEPSEEK-R12026.0180——
SELF-THOUGHTModel=DEEPSEEK-R12026.01801010
SELF-TICKModel=O3-MINI2026.0176.673.336.67
REFLEXModel=DEEPSEEK-R12026.0176.6703.33
SELF-REFINEModel=DEEPSEEK-R12026.0176.673.336.67
SELF-TICKModel=DEEPSEEK-R12026.0160020
REFLEXIONModel=DEEPSEEK-R12026.0156.673.3326.67
SELF-THOUGHTModel=GPT-4O2026.0133.33200
SELF-THOUGHTModel=GPT-4O-MINI2026.013016.676.67
SELF-TICKModel=GPT-4O-MINI2026.0123.336.673.33
INITIAL (t = 0)Model=GPT-4O-MINI2026.0120——
SELF-REFINEModel=GPT-4O2026.0120103.33
SELF-TICKModel=GPT-4O2026.0116.673.330
REFLEXIONModel=GPT-4O2026.0116.67106.67
SELF-REFINEModel=GPT-4O-MINI2026.0113.333.3310
REFLEXIONModel=GPT-4O-MINI2026.0113.336.6713.33
INITIAL (t = 0)Model=GPT-4O2026.0113.33——
REFLEXModel=GPT-4O2026.0113.3300
REFLEXModel=GPT-4O-MINI2026.0110010