Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.
DeepSeek V3.2 (Thinking) wins overall with a score of 69 vs 52 (17 point difference).DeepSeek V3.2 (Thinking) wins 4 out of 4 categories.
DeepSeek V3.2 (Thinking)
84
Mistral 8x7B
62.8
DeepSeek V3.2 (Thinking)
79
Mistral 8x7B
55
DeepSeek V3.2 (Thinking)
86
Mistral 8x7B
64
DeepSeek V3.2 (Thinking)
82
Mistral 8x7B
62
DeepSeek V3.2 (Thinking) scores higher overall with 69 vs 52, a difference of 17 points across all benchmarks.
DeepSeek V3.2 (Thinking) leads in knowledge tasks with an average score of 84 vs 62.8.
DeepSeek V3.2 (Thinking) leads in coding with an average score of 79 vs 55.
DeepSeek V3.2 (Thinking) leads in math with an average score of 86 vs 64.
DeepSeek V3.2 (Thinking) leads in reasoning with an average score of 82 vs 62.