benchlm.ai

Claude Haiku 4.5 vs DeepSeekMath V2

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

Quick Verdict

DeepSeekMath V2 wins overall with a score of 64 vs 57 (7 point difference).DeepSeekMath V2 wins 4 out of 4 categories.

Claude Haiku 4.5

CreatorAnthropic
TypeProprietary
Context200K
Score57

DeepSeekMath V2

CreatorDeepSeek
TypeOpen Weight
Context128K
Score64

Knowledge

DeepSeekMath V2

Claude Haiku 4.5

65.8

DeepSeekMath V2

77.8

68
MMLU
80
67
GPQA
79
65
SuperGPQA
77
63
OpenBookQA
75

Coding

DeepSeekMath V2

Claude Haiku 4.5

60

DeepSeekMath V2

72

60
HumanEval
72

Mathematics

DeepSeekMath V2

Claude Haiku 4.5

67

DeepSeekMath V2

79

68
AIME 2023
80
70
AIME 2024
82
69
AIME 2025
81
64
HMMT Feb 2023
76
66
HMMT Feb 2024
78
65
HMMT Feb 2025
77
67
BRUMO 2025
79

Reasoning

DeepSeekMath V2

Claude Haiku 4.5

64

DeepSeekMath V2

76

65
SimpleQA
77
63
MuSR
75

Frequently Asked Questions

Which is better, Claude Haiku 4.5 or DeepSeekMath V2?

DeepSeekMath V2 scores higher overall with 64 vs 57, a difference of 7 points across all benchmarks.

Which is better for knowledge tasks, Claude Haiku 4.5 or DeepSeekMath V2?

DeepSeekMath V2 leads in knowledge tasks with an average score of 77.8 vs 65.8.

Which is better for coding, Claude Haiku 4.5 or DeepSeekMath V2?

DeepSeekMath V2 leads in coding with an average score of 72 vs 60.

Which is better for math, Claude Haiku 4.5 or DeepSeekMath V2?

DeepSeekMath V2 leads in math with an average score of 79 vs 67.

Which is better for reasoning, Claude Haiku 4.5 or DeepSeekMath V2?

DeepSeekMath V2 leads in reasoning with an average score of 76 vs 64.

Related Comparisons

Claude Haiku 4.5 vs GPT-5.4Claude Haiku 4.5 vs Gemini 3.1 ProClaude Haiku 4.5 vs Claude Opus 4.6Claude Haiku 4.5 vs GPT-5.3 CodexClaude Haiku 4.5 vs Grok 4.1Claude Haiku 4.5 vs GPT-5.2