List your company

Mathematical Reasoning on AIME 2024 (Accuracy and AVG)

86.67AccuracyT3S(R1+235B+QWQ)
71.933275.759179.58583.4109Jan 15, 2026

Evaluation results

MethodLinks
T3S(R1+235B+QWQ)Model=Qwen3-32B2026.0186.6783.34
BaseModel=Qwen3-235B-A22B2026.0185.783.6
T3S(R1+235B+QWQ)Model=Qwen3-8B2026.0184.1780.42
T3S(R1+235B)Model=Qwen3-32B2026.0184.1780.42
T3S(R1+235B)Model=Qwen3-8B2026.0183.3378.33
T3S(R1)Model=Qwen3-32B2026.0183.3378.33
SFT(R1+235B)Model=Qwen3-32B2026.0181.6774.17
BaseModel=Qwen3-32B2026.0181.4676.77
T3S(R1)Model=Qwen3-8B2026.018076.57
BaseModel=DeepSeek-R12026.0179.874.9
SFT(R1+235B+QWQ)Model=Qwen3-32B2026.0178.3373.75
SFT(R1)Model=Qwen3-32B2026.0176.6772.09
BaseModel=Qwen3-8B2026.0175.8371.46
SFT(R1+235B)Model=Qwen3-8B2026.017569.59
SFT(R1+235B+QWQ)Model=Qwen3-8B2026.0173.3369.58
SFT(R1)Model=Qwen3-8B2026.0172.564.17