Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.
DeepSeek V3.2 wins overall with a score of 66 vs 65 (1 point difference).DeepSeek V3.2 wins 3 out of 4 categories.
DeepSeek V3.2
81.8
Qwen2.5-72B
80.8
DeepSeek V3.2
76
Qwen2.5-72B
75
DeepSeek V3.2
83
Qwen2.5-72B
83
DeepSeek V3.2
80
Qwen2.5-72B
79
DeepSeek V3.2 scores higher overall with 66 vs 65, a difference of 1 points across all benchmarks.
DeepSeek V3.2 leads in knowledge tasks with an average score of 81.8 vs 80.8.
DeepSeek V3.2 leads in coding with an average score of 76 vs 75.
DeepSeek V3.2 and Qwen2.5-72B are tied for math with average scores of 83.
DeepSeek V3.2 leads in reasoning with an average score of 80 vs 79.