Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.
Qwen2.5-1M wins overall with a score of 66 vs 24 (42 point difference).Qwen2.5-1M wins 4 out of 4 categories.
DeepSeek V3.1
30.8
Qwen2.5-1M
81.8
DeepSeek V3.1
25
Qwen2.5-1M
76
DeepSeek V3.1
32
Qwen2.5-1M
84
DeepSeek V3.1
30
Qwen2.5-1M
80
Qwen2.5-1M scores higher overall with 66 vs 24, a difference of 42 points across all benchmarks.
Qwen2.5-1M leads in knowledge tasks with an average score of 81.8 vs 30.8.
Qwen2.5-1M leads in coding with an average score of 76 vs 25.
Qwen2.5-1M leads in math with an average score of 84 vs 32.
Qwen2.5-1M leads in reasoning with an average score of 80 vs 30.