Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.
DeepSeekMath V2 wins overall with a score of 64 vs 20 (44 point difference).DeepSeekMath V2 wins 4 out of 4 categories.
DeepSeekMath V2
77.8
Mistral 8x7B v0.2
26.8
DeepSeekMath V2
72
Mistral 8x7B v0.2
21
DeepSeekMath V2
79
Mistral 8x7B v0.2
28
DeepSeekMath V2
76
Mistral 8x7B v0.2
26
DeepSeekMath V2 scores higher overall with 64 vs 20, a difference of 44 points across all benchmarks.
DeepSeekMath V2 leads in knowledge tasks with an average score of 77.8 vs 26.8.
DeepSeekMath V2 leads in coding with an average score of 72 vs 21.
DeepSeekMath V2 leads in math with an average score of 79 vs 28.
DeepSeekMath V2 leads in reasoning with an average score of 76 vs 26.