List your company

Mathematical Reasoning on DM (test)

70.1AccuracyDeepSeekMath-CRPS-60K
44.93251.4665864.534Apr 13, 2026

Evaluation results

MethodLinks
DeepSeekMath-CRPS-60KBackbone=DeepSeekMath-7B, #Samples=60K2026.0470.1
DeepSeekMath-CRPS-30KBackbone=DeepSeekMath-7B, #Samples=30K2026.0468.5
SIGMA-60KBackbone=DeepSeekMath-7B, #Samples=60K2026.0468.4
SIGMA-30KBackbone=DeepSeekMath-7B, #Samples=30K2026.0467.2
DeepSeekMath-CRPS-15KBackbone=DeepSeekMath-7B, #Samples=15K2026.0465.8
MathFusionBackbone=DeepSeekMath-7B, #Samples=60K2026.0465.8
MathFusion (Sequential)Backbone=DeepSeekMath-7B, #Samples=30K2026.0464.6
DeepSeekMath-DARTBackbone=DeepSeekMath-7B, #Samples=60K2026.0462.8
DeepSeekMath-DARTBackbone=DeepSeekMath-7B, #Samples=590K2026.0461.6
DeepSeekMath-RFTBackbone=DeepSeekMath-7B, #Samples=590K2026.0460.2
DeepSeekMath-MMIQCBackbone=DeepSeekMath-7B, #Samples=2.3M2026.0452.9
DeepSeekMath-InstructBackbone=DeepSeekMath-7B, #Samples=780K2026.0452.2
DeepSeekMath-MetaMathBackbone=DeepSeekMath-7B, #Samples=60K2026.0445.9