List your company

Mathematical Reasoning on MATH500 (Avg@5, Pass@5, Cons@5, Tokens)

937.4TokensTRACES
795.8841,751.1172,706.353,661.583Apr 22, 2026

Evaluation results

MethodLinks
TRACESModel=DS-8B, Threshold=0.92026.04937.474.3556.782.428.4
TRACESModel=QwQ-32B, Threshold=0.92026.04942.478.9460.279.837
TRACESModel=DS-14B, Threshold=0.92026.04988.870.8259.579.639.2
TRACESModel=QwQ-32B, Threshold=0.82026.041,258.471.8866.888.242
TRACESModel=DS-8B, Threshold=0.82026.041,290.864.6864.987.436.6
TRACESModel=DS-14B, Threshold=0.82026.041,332.960.6667.58744.8
TRACESModel=DS-14B, Threshold=0.72026.041,713.349.4473.590.250.8
TRACESModel=QwQ-32B, Threshold=0.72026.041,718.261.6174.593.250
TRACESModel=DS-8B, Threshold=0.72026.041,745.252.2571.592.844.2
Basel. P^{(1)}_{system}Model=DS-8B2026.042,139.541.4678.294.252.6
TRACESModel=DS-14B, Threshold=0.62026.042,181.435.6381.195.659.8
Basel. P^{(1)}_{system}Model=DS-14B2026.042,211.434.7487.397.470.8
TRACESModel=DS-8B, Threshold=0.62026.042,264.138.0578.595.253
TRACESModel=QwQ-32B, Threshold=0.62026.042,307.548.4482.496.261.2
Basel. P^{(0)}_{system}Model=DS-14B2026.042,346.230.7788.696.675.4
Basel. P^{(3)}_{system}Model=DS-14B2026.042,53525.1987.996.874.8
Basel. P^{(3)}_{system}Model=DS-8B2026.042,565.329.8178.995.254
Basel. P^{(0)}_{system}Model=DS-8B2026.042,634.427.9281.79659.2
TRACESModel=DS-14B, Threshold=0.52026.042,656.121.6286.597.469.6
Basel. P^{(0)}_{user}Model=DS-14B2026.042,691.520.5893.398.283.4
TRACESModel=DS-8B, Threshold=0.52026.042,725.825.4282.996.861.4
TRACESModel=DS-14B, Threshold=0.42026.042,878.415.0689.297.674.6
Basel. P^{(0)}_{user}Model=QwQ-32B2026.042,908.83595.598.691.6
TRACESModel=QwQ-32B, Threshold=0.52026.042,956.733.9387.89871
Basel. P^{(0)}_{user}Model=DS-8B2026.042,989.618.2186.695.272.2
TRACESModel=DS-8B, Threshold=0.42026.043,049.316.5785.196.866.2
Basel. P^{(1)}_{system}Model=QwQ-32B2026.043,182.428.8992.597.485.6
Basel. P^{(0)}_{system}Model=QwQ-32B2026.043,201.128.4793.297.685.2
StandardModel=DS-14B2026.043,388.8—92.39883.6
TRACESModel=QwQ-32B, Threshold=0.42026.043,395.624.1291.398.678.8
StandardModel=DS-8B2026.043,655—87.89772.6
Basel. P^{(3)}_{system}Model=QwQ-32B2026.043,665.518.0992.697.485.8
StandardModel=QwQ-32B2026.044,475.3—95.498.489.8