List your company

Mathematical Reasoning on GSM8K (Acc, vbal, FP, TN, FN, TP)

97.8AccuracyQwen3
74.50480.55286.692.648Apr 6, 2026

Evaluation results

MethodLinks
Qwen3Model=Qwen32026.0497.865.315.735.592.796.7
Llama4-MaverickModel=Llama4-Maverick2026.0497.26422.342.991.699.1
DeepSeek-V3Model=DeepSeek-V32026.0495.956.97.326.39599.1
Grok-4Model=Grok-42026.0492.762.77.15094.299.7
Tulu3.1-8BModel=Tulu3.1-8B2026.0488.278.850.676.889.898.2
Llama3.1-8BModel=Llama3.1-8B2026.0477.980.459.576.489.396.3
OLMo2-7BModel=OLMo2-7B2026.0475.47646.972.388.696.2