Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.
DeepSeek-R1 wins overall with a score of 35 vs 21 (14 point difference).DeepSeek-R1 wins 4 out of 4 categories.
DeepSeek-R1
41.8
Mistral 7B v0.3
27.8
DeepSeek-R1
36
Mistral 7B v0.3
22
DeepSeek-R1
43
Mistral 7B v0.3
29
DeepSeek-R1
41
Mistral 7B v0.3
27
DeepSeek-R1 scores higher overall with 35 vs 21, a difference of 14 points across all benchmarks.
DeepSeek-R1 leads in knowledge tasks with an average score of 41.8 vs 27.8.
DeepSeek-R1 leads in coding with an average score of 36 vs 22.
DeepSeek-R1 leads in math with an average score of 43 vs 29.
DeepSeek-R1 leads in reasoning with an average score of 41 vs 27.