Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.
DeepSeekMath V2 wins overall with a score of 64 vs 50 (14 point difference).DeepSeekMath V2 wins 4 out of 4 categories.
DeepSeekMath V2
77.8
GPT-4 Turbo
58.5
DeepSeekMath V2
72
GPT-4 Turbo
52
DeepSeekMath V2
79
GPT-4 Turbo
59
DeepSeekMath V2
76
GPT-4 Turbo
57
DeepSeekMath V2 scores higher overall with 64 vs 50, a difference of 14 points across all benchmarks.
DeepSeekMath V2 leads in knowledge tasks with an average score of 77.8 vs 58.5.
DeepSeekMath V2 leads in coding with an average score of 72 vs 52.
DeepSeekMath V2 leads in math with an average score of 79 vs 59.
DeepSeekMath V2 leads in reasoning with an average score of 76 vs 57.