DeepSeek V3.2 (Thinking) vs GPT-5 mini

Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.

Quick Verdict

DeepSeek V3.2 (Thinking) wins overall with a score of 69 vs 68 (1 point difference).DeepSeek V3.2 (Thinking) wins 0 out of 4 categories.

Knowledge

GPT-5 mini

DeepSeek V3.2 (Thinking)

84

GPT-5 mini

85

87
MMLU
88
85
GPQA
86
83
SuperGPQA
84
81
OpenBookQA
82

Coding

GPT-5 mini

DeepSeek V3.2 (Thinking)

79

GPT-5 mini

80

79
HumanEval
80

Mathematics

GPT-5 mini

DeepSeek V3.2 (Thinking)

86

GPT-5 mini

89

87
AIME 2023
90
89
AIME 2024
92
88
AIME 2025
91
83
HMMT Feb 2023
86
85
HMMT Feb 2024
88
84
HMMT Feb 2025
87
86
BRUMO 2025
89

Reasoning

GPT-5 mini

DeepSeek V3.2 (Thinking)

82

GPT-5 mini

83

83
SimpleQA
84
81
MuSR
82

Frequently Asked Questions

Which is better, DeepSeek V3.2 (Thinking) or GPT-5 mini?

DeepSeek V3.2 (Thinking) scores higher overall with 69 vs 68, a difference of 1 points across all benchmarks.

Which is better for knowledge tasks, DeepSeek V3.2 (Thinking) or GPT-5 mini?

GPT-5 mini leads in knowledge tasks with an average score of 85 vs 84.

Which is better for coding, DeepSeek V3.2 (Thinking) or GPT-5 mini?

GPT-5 mini leads in coding with an average score of 80 vs 79.

Which is better for math, DeepSeek V3.2 (Thinking) or GPT-5 mini?

GPT-5 mini leads in math with an average score of 89 vs 86.

Which is better for reasoning, DeepSeek V3.2 (Thinking) or GPT-5 mini?

GPT-5 mini leads in reasoning with an average score of 83 vs 82.