List your company

Mathematical Reasoning on MATH (pv, best, Ac/Y metrics)

87.6Ac (alpha=0.10)R-Q8B
4656.867.678.4May 13, 2026

Evaluation results

MethodLinks
R-Q8BModel=Qwen3-1.7B2026.0587.672.869.277.692.568.7
R-L8BModel=Qwen3-1.7B2026.0587.471.866.274.992.562.9
MVModel=Qwen3-1.7B2026.0587.371.2—71.292.457.9
Ppl-StdModel=Qwen3-1.7B2026.0586.670.863.273.692.662.1
J-Q30BModel=Qwen3-1.7B2026.0586.369.857.57291.759.4
PplModel=Qwen3-1.7B2026.0586.269.863.572.791.661.4
J-G26BModel=Qwen3-1.7B2026.0586705272.191.459
J-R1Q14BModel=Qwen3-1.7B2026.0585.569.154.770.59156.2
OracleModel=Qwen3-1.7B2026.0585.4—85.410085.4100
R-Q8BModel=R1-Qwen-1.5B2026.0582.264.862.255.284.129.6
J-R1Q14BModel=R1-Qwen-1.5B2026.058160.858.350.383.429.5
MVModel=R1-Qwen-1.5B2026.0580.960.3—46.582.824
R-L8BModel=R1-Qwen-1.5B2026.0580.962.357.550.183.327.1
J-G26BModel=R1-Qwen-1.5B2026.0580.961.856.752.383.830
J-Q30BModel=R1-Qwen-1.5B2026.0580.660.652.348.883.129
Ppl-StdModel=R1-Qwen-1.5B2026.0579.459.653.246.582.627.7
PplModel=R1-Qwen-1.5B2026.0577.857.249.842.981.525.1
OracleModel=R1-Qwen-1.5B2026.0576.5—76.510076.5100
J-G26BModel=Gemma-4-E4B2026.0570.358.754.231.172.37.1
R-Q8BModel=Gemma-4-E4B2026.0570.260.957.732.264.67.5
J-R1Q14BModel=Gemma-4-E4B2026.0569.258.651.530.777.76.9
R-L8BModel=Gemma-4-E4B2026.0569.158.757.131.572.79.5
OracleModel=Gemma-4-E4B2026.0568.5—68.510068.5100
MVModel=Gemma-4-E4B2026.0568.558.3—21.1—0
PplModel=Gemma-4-E4B2026.0567.951.146.829.184.16.1
Ppl-StdModel=Gemma-4-E4B2026.0567.850.645.929.167.37.7
J-Q30BModel=Gemma-4-E4B2026.0567.456.851.228.152.91.6
GreedyModel=Qwen3-1.7B2026.0555.5——10055.5100
GreedyModel=Gemma-4-E4B2026.0552.1——10052.1100
GreedyModel=R1-Qwen-1.5B2026.0547.6——10047.6100