Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.
Both models are tied with an overall score of 66.
DeepSeek V3.2
81.8
Qwen2.5-1M
81.8
DeepSeek V3.2
76
Qwen2.5-1M
76
DeepSeek V3.2
83
Qwen2.5-1M
84
DeepSeek V3.2
80
Qwen2.5-1M
80
DeepSeek V3.2 and Qwen2.5-1M are tied with identical overall scores of 66.
DeepSeek V3.2 and Qwen2.5-1M are tied for knowledge tasks with average scores of 81.8.
DeepSeek V3.2 and Qwen2.5-1M are tied for coding with average scores of 76.
Qwen2.5-1M leads in math with an average score of 84 vs 83.
DeepSeek V3.2 and Qwen2.5-1M are tied for reasoning with average scores of 80.