Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.
DeepSeekMath V2 wins overall with a score of 64 vs 45 (19 point difference).DeepSeekMath V2 wins 4 out of 4 categories.
DeepSeekMath V2
77.8
Nemotron-4 15B
51.8
DeepSeekMath V2
72
Nemotron-4 15B
46
DeepSeekMath V2
79
Nemotron-4 15B
53
DeepSeekMath V2
76
Nemotron-4 15B
51
DeepSeekMath V2 scores higher overall with 64 vs 45, a difference of 19 points across all benchmarks.
DeepSeekMath V2 leads in knowledge tasks with an average score of 77.8 vs 51.8.
DeepSeekMath V2 leads in coding with an average score of 72 vs 46.
DeepSeekMath V2 leads in math with an average score of 79 vs 53.
DeepSeekMath V2 leads in reasoning with an average score of 76 vs 51.