Claude Haiku 4.5 vs DeepSeek V3.2

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

Quick Verdict

DeepSeek V3.2 wins overall with a score of 66 vs 57 (9 point difference).DeepSeek V3.2 wins 4 out of 4 categories.

Knowledge

DeepSeek V3.2

Claude Haiku 4.5

65.8

DeepSeek V3.2

81.8

68
MMLU
84
67
GPQA
83
65
SuperGPQA
81
63
OpenBookQA
79

Coding

DeepSeek V3.2

Claude Haiku 4.5

60

DeepSeek V3.2

76

60
HumanEval
76

Mathematics

DeepSeek V3.2

Claude Haiku 4.5

67

DeepSeek V3.2

83

68
AIME 2023
84
70
AIME 2024
86
69
AIME 2025
85
64
HMMT Feb 2023
80
66
HMMT Feb 2024
82
65
HMMT Feb 2025
81
67
BRUMO 2025
83

Reasoning

DeepSeek V3.2

Claude Haiku 4.5

64

DeepSeek V3.2

80

65
SimpleQA
81
63
MuSR
79

Frequently Asked Questions

Which is better, Claude Haiku 4.5 or DeepSeek V3.2?

DeepSeek V3.2 scores higher overall with 66 vs 57, a difference of 9 points across all benchmarks.

Which is better for knowledge tasks, Claude Haiku 4.5 or DeepSeek V3.2?

DeepSeek V3.2 leads in knowledge tasks with an average score of 81.8 vs 65.8.

Which is better for coding, Claude Haiku 4.5 or DeepSeek V3.2?

DeepSeek V3.2 leads in coding with an average score of 76 vs 60.

Which is better for math, Claude Haiku 4.5 or DeepSeek V3.2?

DeepSeek V3.2 leads in math with an average score of 83 vs 67.

Which is better for reasoning, Claude Haiku 4.5 or DeepSeek V3.2?

DeepSeek V3.2 leads in reasoning with an average score of 80 vs 64.