Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.
DeepSeek-R1 wins overall with a score of 35 vs 22 (13 point difference).DeepSeek-R1 wins 4 out of 4 categories.
DeepSeek-R1
41.8
GPT-OSS 20B
28.8
DeepSeek-R1
36
GPT-OSS 20B
23
DeepSeek-R1
43
GPT-OSS 20B
30
DeepSeek-R1
41
GPT-OSS 20B
28
DeepSeek-R1 scores higher overall with 35 vs 22, a difference of 13 points across all benchmarks.
DeepSeek-R1 leads in knowledge tasks with an average score of 41.8 vs 28.8.
DeepSeek-R1 leads in coding with an average score of 36 vs 23.
DeepSeek-R1 leads in math with an average score of 43 vs 30.
DeepSeek-R1 leads in reasoning with an average score of 41 vs 28.