benchlm.ai

Claude 3 Haiku 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 46 (23 point difference).DeepSeek V3.2 (Thinking) wins 4 out of 4 categories.

Claude 3 Haiku

CreatorAnthropic
TypeProprietary
Context200K
Score46

DeepSeek V3.2 (Thinking)

CreatorDeepSeek
TypeOpen Weight
Context128K
Score69

Knowledge

DeepSeek V3.2 (Thinking)

Claude 3 Haiku

54.5

DeepSeek V3.2 (Thinking)

84

56
MMLU
87
56
GPQA
85
54
SuperGPQA
83
52
OpenBookQA
81

Coding

DeepSeek V3.2 (Thinking)

Claude 3 Haiku

48

DeepSeek V3.2 (Thinking)

79

48
HumanEval
79

Mathematics

DeepSeek V3.2 (Thinking)

Claude 3 Haiku

55

DeepSeek V3.2 (Thinking)

86

56
AIME 2023
87
58
AIME 2024
89
57
AIME 2025
88
52
HMMT Feb 2023
83
54
HMMT Feb 2024
85
53
HMMT Feb 2025
84
55
BRUMO 2025
86

Reasoning

DeepSeek V3.2 (Thinking)

Claude 3 Haiku

53

DeepSeek V3.2 (Thinking)

82

54
SimpleQA
83
52
MuSR
81

Frequently Asked Questions

Which is better, Claude 3 Haiku or DeepSeek V3.2 (Thinking)?

DeepSeek V3.2 (Thinking) scores higher overall with 69 vs 46, a difference of 23 points across all benchmarks.

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

DeepSeek V3.2 (Thinking) leads in knowledge tasks with an average score of 84 vs 54.5.

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

DeepSeek V3.2 (Thinking) leads in coding with an average score of 79 vs 48.

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

DeepSeek V3.2 (Thinking) leads in math with an average score of 86 vs 55.

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

DeepSeek V3.2 (Thinking) leads in reasoning with an average score of 82 vs 53.

Related Comparisons

Claude 3 Haiku vs GPT-5.4Claude 3 Haiku vs Gemini 3.1 ProClaude 3 Haiku vs Claude Opus 4.6Claude 3 Haiku vs GPT-5.3 CodexClaude 3 Haiku vs Grok 4.1Claude 3 Haiku vs GPT-5.2