Claude 4.5 Haiku (Reasoning) vs DeepSeek R1 Distill Qwen 32B | grube.ai

Back to Benchmarks

Claude 4.5 Haiku (Reasoning)

Anthropic

vs

DeepSeek R1 Distill Qwen 32B

DeepSeekvia NextBit

Claude 4.5 Haiku (Reasoning) wins 9 of 11

Index Scores

37.1

+19.9

Intelligence

17.2

32.6

+32.6

Coding

0.0

83.7

+20.7

Math

63.0

Benchmarks

76.0

+2.1

MMLU Pro

73.9

67.2

+5.7

GPQA

61.5

9.7

+4.2

HLE

5.5

61.5

+34.5

LiveCode

27.0

43.3

+5.7

SciCode

37.6

0.0

MATH-500

+94.1

94.1

0.0

AIME

+68.7

68.7

Speed

66

+42

TPS

24

Shared

Winning delta

crafted by bart stefanski