benchlm.ai

DeepSeek V3.1 vs DeepSeek V3.2 (Thinking)

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 24 (45 point difference).DeepSeek V3.2 (Thinking) wins 4 out of 4 categories.

DeepSeek V3.1

CreatorDeepSeek
TypeOpen Weight
Context128K
Score24

DeepSeek V3.2 (Thinking)

CreatorDeepSeek
TypeOpen Weight
Context128K
Score69

Knowledge

DeepSeek V3.2 (Thinking)

DeepSeek V3.1

30.8

DeepSeek V3.2 (Thinking)

84

33
MMLU
87
32
GPQA
85
30
SuperGPQA
83
28
OpenBookQA
81

Coding

DeepSeek V3.2 (Thinking)

DeepSeek V3.1

25

DeepSeek V3.2 (Thinking)

79

25
HumanEval
79

Mathematics

DeepSeek V3.2 (Thinking)

DeepSeek V3.1

32

DeepSeek V3.2 (Thinking)

86

33
AIME 2023
87
35
AIME 2024
89
34
AIME 2025
88
29
HMMT Feb 2023
83
31
HMMT Feb 2024
85
30
HMMT Feb 2025
84
32
BRUMO 2025
86

Reasoning

DeepSeek V3.2 (Thinking)

DeepSeek V3.1

30

DeepSeek V3.2 (Thinking)

82

31
SimpleQA
83
29
MuSR
81

Frequently Asked Questions

Which is better, DeepSeek V3.1 or DeepSeek V3.2 (Thinking)?

DeepSeek V3.2 (Thinking) scores higher overall with 69 vs 24, a difference of 45 points across all benchmarks.

Which is better for knowledge tasks, DeepSeek V3.1 or DeepSeek V3.2 (Thinking)?

DeepSeek V3.2 (Thinking) leads in knowledge tasks with an average score of 84 vs 30.8.

Which is better for coding, DeepSeek V3.1 or DeepSeek V3.2 (Thinking)?

DeepSeek V3.2 (Thinking) leads in coding with an average score of 79 vs 25.

Which is better for math, DeepSeek V3.1 or DeepSeek V3.2 (Thinking)?

DeepSeek V3.2 (Thinking) leads in math with an average score of 86 vs 32.

Which is better for reasoning, DeepSeek V3.1 or DeepSeek V3.2 (Thinking)?

DeepSeek V3.2 (Thinking) leads in reasoning with an average score of 82 vs 30.

Related Comparisons

DeepSeek V3.1 vs GPT-5.4DeepSeek V3.1 vs Gemini 3.1 ProDeepSeek V3.1 vs Claude Opus 4.6DeepSeek V3.1 vs GPT-5.3 CodexDeepSeek V3.1 vs Grok 4.1DeepSeek V3.1 vs GPT-5.2