Claude 3.5 Sonnet 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 55 (14 point difference).DeepSeek V3.2 (Thinking) wins 4 out of 4 categories.

Knowledge

DeepSeek V3.2 (Thinking)

Claude 3.5 Sonnet

63.5

DeepSeek V3.2 (Thinking)

84

65
MMLU
87
65
GPQA
85
63
SuperGPQA
83
61
OpenBookQA
81

Coding

DeepSeek V3.2 (Thinking)

Claude 3.5 Sonnet

57

DeepSeek V3.2 (Thinking)

79

57
HumanEval
79

Mathematics

DeepSeek V3.2 (Thinking)

Claude 3.5 Sonnet

64

DeepSeek V3.2 (Thinking)

86

65
AIME 2023
87
67
AIME 2024
89
66
AIME 2025
88
61
HMMT Feb 2023
83
63
HMMT Feb 2024
85
62
HMMT Feb 2025
84
64
BRUMO 2025
86

Reasoning

DeepSeek V3.2 (Thinking)

Claude 3.5 Sonnet

62

DeepSeek V3.2 (Thinking)

82

63
SimpleQA
83
61
MuSR
81

Frequently Asked Questions

Which is better, Claude 3.5 Sonnet or DeepSeek V3.2 (Thinking)?

DeepSeek V3.2 (Thinking) scores higher overall with 69 vs 55, a difference of 14 points across all benchmarks.

Which is better for knowledge tasks, Claude 3.5 Sonnet or DeepSeek V3.2 (Thinking)?

DeepSeek V3.2 (Thinking) leads in knowledge tasks with an average score of 84 vs 63.5.

Which is better for coding, Claude 3.5 Sonnet or DeepSeek V3.2 (Thinking)?

DeepSeek V3.2 (Thinking) leads in coding with an average score of 79 vs 57.

Which is better for math, Claude 3.5 Sonnet or DeepSeek V3.2 (Thinking)?

DeepSeek V3.2 (Thinking) leads in math with an average score of 86 vs 64.

Which is better for reasoning, Claude 3.5 Sonnet or DeepSeek V3.2 (Thinking)?

DeepSeek V3.2 (Thinking) leads in reasoning with an average score of 82 vs 62.