Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.
DeepSeekMath V2 wins overall with a score of 64 vs 24 (40 point difference).DeepSeekMath V2 wins 4 out of 4 categories.
DeepSeek V3.1
30.8
DeepSeekMath V2
77.8
DeepSeek V3.1
25
DeepSeekMath V2
72
DeepSeek V3.1
32
DeepSeekMath V2
79
DeepSeek V3.1
30
DeepSeekMath V2
76
DeepSeekMath V2 scores higher overall with 64 vs 24, a difference of 40 points across all benchmarks.
DeepSeekMath V2 leads in knowledge tasks with an average score of 77.8 vs 30.8.
DeepSeekMath V2 leads in coding with an average score of 72 vs 25.
DeepSeekMath V2 leads in math with an average score of 79 vs 32.
DeepSeekMath V2 leads in reasoning with an average score of 76 vs 30.