Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.
DeepSeekMath V2 wins overall with a score of 64 vs 55 (9 point difference).DeepSeekMath V2 wins 4 out of 4 categories.
DeepSeekMath V2
77.8
Nemotron 3 Super 100B
62.8
DeepSeekMath V2
72
Nemotron 3 Super 100B
57
DeepSeekMath V2
79
Nemotron 3 Super 100B
64
DeepSeekMath V2
76
Nemotron 3 Super 100B
61
DeepSeekMath V2 scores higher overall with 64 vs 55, a difference of 9 points across all benchmarks.
DeepSeekMath V2 leads in knowledge tasks with an average score of 77.8 vs 62.8.
DeepSeekMath V2 leads in coding with an average score of 72 vs 57.
DeepSeekMath V2 leads in math with an average score of 79 vs 64.
DeepSeekMath V2 leads in reasoning with an average score of 76 vs 61.