Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.
DeepSeek-R1 wins overall with a score of 35 vs 32 (3 point difference).DeepSeek-R1 wins 4 out of 4 categories.
DeepSeek-R1
41.8
Nova Pro
38.8
DeepSeek-R1
36
Nova Pro
33
DeepSeek-R1
43
Nova Pro
40
DeepSeek-R1
41
Nova Pro
38
DeepSeek-R1 scores higher overall with 35 vs 32, a difference of 3 points across all benchmarks.
DeepSeek-R1 leads in knowledge tasks with an average score of 41.8 vs 38.8.
DeepSeek-R1 leads in coding with an average score of 36 vs 33.
DeepSeek-R1 leads in math with an average score of 43 vs 40.
DeepSeek-R1 leads in reasoning with an average score of 41 vs 38.