Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.
DeepSeek V3.2 (Thinking) wins overall with a score of 69 vs 62 (7 point difference).DeepSeek V3.2 (Thinking) wins 4 out of 4 categories.
DeepSeek V3.2 (Thinking)
84
Kimi K2.5
74.8
DeepSeek V3.2 (Thinking)
79
Kimi K2.5
69
DeepSeek V3.2 (Thinking)
86
Kimi K2.5
76
DeepSeek V3.2 (Thinking)
82
Kimi K2.5
73
DeepSeek V3.2 (Thinking) scores higher overall with 69 vs 62, a difference of 7 points across all benchmarks.
DeepSeek V3.2 (Thinking) leads in knowledge tasks with an average score of 84 vs 74.8.
DeepSeek V3.2 (Thinking) leads in coding with an average score of 79 vs 69.
DeepSeek V3.2 (Thinking) leads in math with an average score of 86 vs 76.
DeepSeek V3.2 (Thinking) leads in reasoning with an average score of 82 vs 73.