benchlm.ai

DeepSeek V3.2 (Thinking) vs Gemini 3 Pro

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

Quick Verdict

Gemini 3 Pro wins overall with a score of 78 vs 69 (9 point difference).Gemini 3 Pro wins 4 out of 4 categories.

DeepSeek V3.2 (Thinking)

CreatorDeepSeek
TypeOpen Weight
Context128K
Score69

Gemini 3 Pro

CreatorGoogle
TypeProprietary
Context2M
Score78

Knowledge

Gemini 3 Pro

DeepSeek V3.2 (Thinking)

84

Gemini 3 Pro

96

87
MMLU
99
85
GPQA
97
83
SuperGPQA
95
81
OpenBookQA
93

Coding

Gemini 3 Pro

DeepSeek V3.2 (Thinking)

79

Gemini 3 Pro

91

79
HumanEval
91

Mathematics

Gemini 3 Pro

DeepSeek V3.2 (Thinking)

86

Gemini 3 Pro

97.1

87
AIME 2023
99
89
AIME 2024
99
88
AIME 2025
98
83
HMMT Feb 2023
95
85
HMMT Feb 2024
97
84
HMMT Feb 2025
96
86
BRUMO 2025
96

Reasoning

Gemini 3 Pro

DeepSeek V3.2 (Thinking)

82

Gemini 3 Pro

94

83
SimpleQA
95
81
MuSR
93

Frequently Asked Questions

Which is better, DeepSeek V3.2 (Thinking) or Gemini 3 Pro?

Gemini 3 Pro scores higher overall with 78 vs 69, a difference of 9 points across all benchmarks.

Which is better for knowledge tasks, DeepSeek V3.2 (Thinking) or Gemini 3 Pro?

Gemini 3 Pro leads in knowledge tasks with an average score of 96 vs 84.

Which is better for coding, DeepSeek V3.2 (Thinking) or Gemini 3 Pro?

Gemini 3 Pro leads in coding with an average score of 91 vs 79.

Which is better for math, DeepSeek V3.2 (Thinking) or Gemini 3 Pro?

Gemini 3 Pro leads in math with an average score of 97.1 vs 86.

Which is better for reasoning, DeepSeek V3.2 (Thinking) or Gemini 3 Pro?

Gemini 3 Pro leads in reasoning with an average score of 94 vs 82.

Related Comparisons

DeepSeek V3.2 (Thinking) vs GPT-5.4DeepSeek V3.2 (Thinking) vs Gemini 3.1 ProDeepSeek V3.2 (Thinking) vs Claude Opus 4.6DeepSeek V3.2 (Thinking) vs GPT-5.3 CodexDeepSeek V3.2 (Thinking) vs Grok 4.1DeepSeek V3.2 (Thinking) vs GPT-5.2