Claude 4.5 Haiku (Reasoning) vs GPT-4.1 mini | grube.ai

Back to Benchmarks

Claude 4.5 Haiku (Reasoning)

Anthropic

vs

GPT-4.1 mini

OpenAI

Claude 4.5 Haiku (Reasoning) wins 8 of 11

Index Scores

37.1

+14.2

Intelligence

22.9

32.6

+14.1

Coding

18.5

83.7

+37.4

Math

46.3

Benchmarks

76.0

MMLU Pro

+2.1

78.1

67.2

+0.8

GPQA

66.4

9.7

+5.1

HLE

4.6

61.5

+13.2

LiveCode

48.3

43.3

+2.9

SciCode

40.4

0.0

MATH-500

+92.5

92.5

0.0

AIME

+43.0

43.0

Speed

66

+15

TPS

51

Shared

Winning delta

crafted by bart stefanski