List your company

Common-sense Reasoning Suite (BoolQ, SciQ, PIQA, WinoG., ARC-C, HellaS.)

88BoolQ AccuracyDense
35.79249.34662.976.454Jun 1, 2026

Evaluation results

MethodLinks
DenseBase Model=Qwen2.5 7B, Active Params=7.62B, #FT Tokens=18T*2026.068896.678.676.264.180.380.6
DenseBase Model=LLaMA-3 8B, Active Params=8.03B, #FT Tokens=15T*2026.0683.296.279.677.358.382.179.4
DenseBase Model=LLaMA-2 7B, Active Params=6.74B, #FT Tokens=2T*2026.06829478.174.352.578.976.6
DOT-MoEBase Model=Qwen2.5 7B, Active Params=4.76B, #FT Tokens=1.2B2026.0681.394.673.168.149.867.373.4
LLaMA-MoE-v2 (E8A1S1)Base Model=LLaMA-3 8B, Active Params=3.80B, #FT Tokens=7B2026.0676.992.867.958.440.253.765
DOT-MoEBase Model=LLaMA-3 8B, Active Params=3.80B, #FT Tokens=7B2026.0675.496.273.366.149.16671
DOT-MoEBase Model=Qwen2.5 7B, Active Params=3.34B, #FT Tokens=1.2B2026.0675.29570.461.342.858.467.2
DOT-MoEBase Model=LLaMA-3 8B, Active Params=3.80B, #FT Tokens=1.2B2026.067594.270.263.842.461.167.8
LLaMA-MoE-v2 (E8A2)Base Model=LLaMA-3 8B, Active Params=3.80B, #FT Tokens=7B2026.0674.694.569.360.542.85966.8
DOT-MoEBase Model=LLaMA-2 7B, Active Params=3.49B, #FT Tokens=1.2B2026.0672.594.369.362.540.960.266.6
CMoE (E8A1S1)Base Model=LLaMA-3 8B, Active Params=3.80B, #FT Tokens=1.2B2026.0671.194.469.559.538.255.364.7
CMoE (E8A1S1)Base Model=Qwen2.5 7B, Active Params=3.34B, #FT Tokens=1.2B2026.0669.493.867.65937.857.264.1
CMoE (E8S1A1)Base Model=LLaMA-2 7B, Active Params=3.49B, #FT Tokens=1.2B2026.065577.557.154.127.638.851.7
LLaMA-MoE-v2 (E8A2)Base Model=LLaMA-2 7B, Active Params=3.49B, #FT Tokens=1.2B2026.0651.36756.652.925.735.148.1
LLaMA-MoE (E8A2)Base Model=LLaMA-2 7B, Active Params=3.49B, #FT Tokens=1.2B2026.0637.82049.750.125.826.234.9